Why You're Not Actually Productive...Dark Side Of Productivity Nobody Talks About

00:00:00.700 | So I recently heard a term that I really liked on the Chris Williamson podcast.

00:00:05.900 | The term he used was productivity rain dance.

00:00:10.100 | So I think this idea actually gets at something critical,

00:00:14.100 | a critical concept about work and productivity and technology.

00:00:16.900 | It's a concept that I think is exploring because it might explain

00:00:19.800 | some of both your frustrations and confusions

00:00:23.200 | about trying to produce stuff that matters.

00:00:26.500 | So what we'll do is I'm going to load up the clip.

00:00:29.400 | So we'll hear Chris actually talking about it himself.

00:00:31.800 | Then we'll get into a little more detail what he's talking about,

00:00:34.700 | why I think it's a problem,

00:00:35.400 | and then give some advice for how to get around it.

00:00:37.200 | All right, so let's start with the clip.

00:00:38.700 | This comes from, I think he was interviewing,

00:00:41.000 | Chris Williamson was interviewing Sahil Bloom,

00:00:42.800 | but the person talking here in the clip is Chris himself.

00:00:45.500 | So let's load up that audio.

00:00:47.500 | Look, I come from a productivity background.

00:00:51.800 | When I first started the show, I was chatting shit about

00:00:54.500 | Pomodoro timers and Notion external brains

00:00:57.900 | and Ebbinghaus forgetting curves and all of that.

00:01:00.800 | I've been through the ringer, which is why I'm allowed to say it.

00:01:02.800 | And you realize after a while

00:01:07.200 | that it ends up being this weird superstitious rain dance you're doing,

00:01:10.400 | this sort of odd sort of productivity rain dance

00:01:13.300 | in the desperate hope that later that day

00:01:15.800 | you're going to get something done.

00:01:17.200 | And some of that stuff does really help.

00:01:19.100 | And you kind of need to go through this process of,

00:01:21.900 | "Ah, it wasn't the 15 push-ups before I do my calls.

00:01:26.300 | Oh, it wasn't, it was this thing."

00:01:28.200 | That's the highest point of learning.

00:01:29.300 | All right, so that's the clip I wanted to point out there.

00:01:32.200 | That studio, I've been there, Jesse.

00:01:33.600 | That's one of the studios Chris uses in Austin.

00:01:36.800 | I did a show last spring.

00:01:38.300 | It has the two weird fluorescent lights

00:01:40.400 | that look like they're coming out of you.

00:01:42.300 | So I love that idea, this notion of productivity rain dance.

00:01:46.600 | So I started looking into it.

00:01:47.700 | Chris has talked about it before.

00:01:50.100 | So then I found this post,

00:01:51.500 | which I'll load up on the screen here

00:01:52.700 | for people who are watching instead of just listening.

00:01:55.600 | This post is from last summer.

00:01:57.200 | And he elaborates more on this idea.

00:01:59.900 | I'm going to read some from this post

00:02:01.200 | because I think it's going to help us get closer

00:02:02.800 | to what the key idea is here.

00:02:04.100 | So in this post, Chris says,

00:02:06.500 | "During my interview review, I asked myself two new questions.

00:02:10.000 | What do I do that I think is productive but isn't?

00:02:13.300 | What do I do that I don't think is productive but actually is?

00:02:16.600 | These were surprisingly easy to work out.

00:02:18.600 | Sitting at my desk when I'm not working,

00:02:21.100 | being on calls with no actual objective,

00:02:22.700 | keeping Slack notifications at zero,

00:02:24.700 | sitting on email trying to get the unread number down,

00:02:27.100 | saying yes to a random dinner

00:02:29.400 | when someone is coming through town,

00:02:30.800 | organizing meetups with friends from different social groups,

00:02:33.200 | walking around without anything in my ears,

00:02:35.500 | reading, visiting new places.

00:02:36.600 | After six months of reflecting on my answers,

00:02:38.500 | I realized I had a fundamental oversight.

00:02:40.400 | I hadn't been properly linking inputs to outcomes.

00:02:43.800 | I had basically created a productivity rain dance."

00:02:48.000 | All right, I think this elaboration is useful

00:02:52.500 | because if we think about the various examples

00:02:55.400 | that Chris gave here talking about these rain dances,

00:02:57.600 | they seem sort of different, right?

00:03:00.600 | We had on one hand him talking about things

00:03:03.000 | like external brains or Ebenhaus forgetting curves,

00:03:07.400 | these sort of overly wrought productivity systems

00:03:11.300 | or tools you're trying to build.

00:03:12.600 | But here he's also talking about things

00:03:14.400 | like just being in his email inbox too long

00:03:17.200 | or like trying to get the Slacks all cleared out

00:03:19.800 | or just sitting at his desk like sort of acting,

00:03:21.900 | like playing at working because it sort of feels productive.

00:03:24.600 | He is somehow unifying all of that

00:03:27.700 | under the same term of productivity rain dance.

00:03:30.600 | And the way he does it and the way I want to focus on here

00:03:33.100 | is by saying what unifies all those examples

00:03:35.600 | is focusing on input instead of output.

00:03:38.900 | So when you focus on input,

00:03:41.600 | you're focusing on activity,

00:03:44.400 | you're focusing on the potential for future activity.

00:03:47.800 | So sitting at your desk checking email,

00:03:49.600 | you're doing something.

00:03:50.800 | The productivity, I'm putting lots of inputs

00:03:53.500 | into my productivity equation, that must be good.

00:03:55.300 | When you're doing something

00:03:56.500 | like building a complicated productivity system,

00:04:00.500 | again, this is like inputs.

00:04:01.800 | I'm working on my productivity function.

00:04:03.300 | I'm building up my opportunities

00:04:07.000 | for finding information or getting things done.

00:04:09.000 | But what you're not looking at is the output.

00:04:10.700 | What am I actually producing that matters?

00:04:13.500 | And am I producing enough of it?

00:04:17.100 | So when you look around the modern office environment

00:04:19.300 | and see everyone frantically answering emails

00:04:21.500 | as they jump on and off Zoom meetings

00:04:23.300 | or watch the solo entrepreneur lose a morning

00:04:26.300 | to optimizing their chat GPT-powered personalized assistant,

00:04:30.600 | you're observing rain dances.

00:04:32.700 | Everyone's busy, but no one is asking

00:04:36.000 | if all these gyrations are actually opening the clouds.

00:04:39.900 | So the danger with rain dances is that they are easier

00:04:46.900 | than the actual work of producing stuff.

00:04:48.600 | It's easier to sit at your desk and jump on calls

00:04:52.700 | or answer emails than it is to actually do something hard.

00:04:55.600 | It's easier to work on your chat GPT-powered assistant.

00:05:00.700 | It's kind of fun.

00:05:01.300 | You're watching YouTube videos and scripting things together.

00:05:04.000 | That's easier than actually writing the thing

00:05:07.000 | that the assistant was going to help you with.

00:05:08.500 | And because it's easier, we tend to gravitate towards that.

00:05:12.300 | We want to spend much more time on it.

00:05:14.800 | We're going to gravitate towards what's easier

00:05:16.700 | versus what's harder if we're not differentiating

00:05:18.400 | between inputs and outputs.

00:05:19.600 | And soon you get yourself to a point where you're super busy,

00:05:23.900 | but very little is actually getting done.

00:05:26.600 | Here's a quote from William Sim on this.

00:05:28.100 | "Obsessing over process while being detached from outcomes

00:05:31.900 | gives you all the pain of hard work

00:05:33.800 | with none of the actual results."

00:05:35.200 | So this is what a productivity rain dance is.

00:05:38.900 | It's when you focus on inputs and ignore outputs.

00:05:41.000 | And the reason why they're dangerous

00:05:43.800 | is that they're more fun and satisfying

00:05:46.500 | or easy to get started with in the moment

00:05:48.100 | than the actual hard work.

00:05:49.200 | And so paradoxically, the stuff you produce gets worse.

00:05:53.700 | So what is the right response to this?

00:05:56.900 | Because I think a lot of people

00:05:58.000 | feel the frustration of their productivity rain dances

00:06:01.500 | without realizing what that frustration is coming from.

00:06:03.700 | And there's two responses that I think are natural

00:06:06.300 | that I both think are flawed.

00:06:07.500 | We hear these often, though,

00:06:09.500 | in sort of online discussions of productivity culture.

00:06:13.100 | The first response is to just demonize work itself.

00:06:15.500 | Like, I don't know, I'm doing stuff all the time.

00:06:17.900 | I'm feeling exhausted.

00:06:19.000 | Not much is being produced.

00:06:20.800 | Maybe the problem is like work itself is kind of meaningless.

00:06:23.500 | It's constructed.

00:06:24.900 | It's like a mirage of late-stage capitalism.

00:06:27.900 | It's all just hustle culture.

00:06:29.200 | So you could just try to demonize work itself

00:06:31.000 | and lean into some sort of like quiet quitting type mentality.

00:06:33.400 | But this is not going to solve your problem.

00:06:35.700 | Just doing your work worse

00:06:38.200 | or just doing overall a lot less work

00:06:40.100 | is going to just get you into other types of problems.

00:06:42.200 | Your business is going to falter.

00:06:43.600 | Your boss is going to move on from you.

00:06:45.400 | The other common response to this is to say,

00:06:47.500 | ah, the problem here is just thinking about productivity itself.

00:06:51.300 | So let's just get rid of all attempts to organize my efforts.

00:06:53.800 | That's the problem.

00:06:54.400 | But this becomes a problem, too,

00:06:57.100 | because if you have no attempts to organize your efforts,

00:06:59.700 | other people organize them for you.

00:07:01.500 | Your life will just become this exhausted slurry

00:07:04.200 | of answering other people's requests

00:07:05.600 | and trying to make other people's lives easier.

00:07:07.200 | So we still need some organization.

00:07:10.200 | So what is the right response

00:07:12.000 | to the productivity rain dance phenomenon?

00:07:14.200 | I think it is, as Williamson suggests,

00:07:17.800 | to turn your attention from inputs to outputs.

00:07:20.400 | To identify the most valuable thing you do in your job

00:07:24.000 | and then figure out

00:07:25.400 | what helps you do that better.

00:07:28.200 | And this should be what matters.

00:07:31.000 | Above all else, this should be where your focus is.

00:07:33.400 | Now, here's the thing.

00:07:34.600 | The answers to these questions,

00:07:35.900 | like what really matters

00:07:38.000 | and what is actually helping me do that thing better,

00:07:41.000 | you know, at a higher level of quality and output,

00:07:43.600 | the answers to those questions aren't necessarily simple.

00:07:47.200 | But they're also different.

00:07:49.600 | The things you come up here are different

00:07:51.400 | than what productivity rain dances will produce.

00:07:54.200 | They're different because they're not symbolic.

00:07:56.400 | They're different because they're not

00:07:57.800 | busyness for the sake of busyness.

00:07:59.400 | They're instead focused on clear, measurable goals

00:08:01.600 | of producing more results that matter.

00:08:04.700 | So when you start focusing on outputs,

00:08:07.400 | you begin to build an approach to productivity

00:08:12.000 | that just works and it's not exciting

00:08:15.200 | and it's hard and it's simple

00:08:19.100 | and it's probably, you know,

00:08:20.900 | there's only so much YouTube content you could get out of it,

00:08:23.500 | but it's the stuff that actually works.

00:08:25.200 | Hey, it's Cal.

00:08:26.100 | I wanted to interrupt briefly to say that

00:08:28.200 | if you're enjoying this video,

00:08:29.800 | then you need to check out my new book,

00:08:32.100 | Slow Productivity,

00:08:34.100 | The Lost Art of Accomplishment Without Burnout.

00:08:37.600 | This is like the Bible for most of the ideas

00:08:40.600 | we talk about here in these videos.

00:08:43.200 | You can get a free excerpt at calnewport.com/slow.

00:08:48.500 | I know you're going to like it.

00:08:50.200 | Check it out.

00:08:51.300 | Now, let's get back to the video.

00:08:52.600 | So let me give you a couple of specific examples here.

00:08:54.500 | What are the types of less flashy,

00:08:56.200 | get-it-done type things that show up

00:08:58.800 | when you start asking

00:08:59.600 | what actually gets the important stuff done?

00:09:01.800 | You find things like work quotas.

00:09:04.500 | All right, for each of these type of things

00:09:07.800 | I am expected to do in my job,

00:09:09.500 | I have a quota on how many I do at the same time.

00:09:12.000 | Why?

00:09:12.800 | Because if I have too many things going on,

00:09:14.700 | I get overloaded.

00:09:15.900 | If I get overloaded,

00:09:16.700 | I can't do the hard stuff well.

00:09:18.600 | I want to do the hard stuff well,

00:09:20.000 | so I only take on one project of this at a time.

00:09:23.000 | I only do this many committees at a time.

00:09:24.400 | I only do this many pay-per-views at a time.

00:09:26.000 | It's not exciting.

00:09:27.200 | There's no AI involved.

00:09:28.900 | There's no cool notebook and a new pen involved,

00:09:31.500 | but it works.

00:09:32.300 | Separating active versus waiting projects.

00:09:35.400 | This also works.

00:09:36.200 | A big concept from my book Slow Productivity.

00:09:38.300 | Here's the things you have to do.

00:09:40.800 | Just look at the first few

00:09:42.400 | and say I'm actively working on these

00:09:43.800 | and everything else is in a waiting state.

00:09:45.700 | And if you ask me about it,

00:09:46.600 | I'll say it's in a waiting state

00:09:47.700 | and I'll tell you as soon as it gets to active.

00:09:49.300 | But as long as it's in a waiting state,

00:09:50.800 | I'm not doing meetings or emails or calls about it.

00:09:53.000 | This prevents the totality of the things

00:09:56.100 | that you have to do

00:09:57.100 | generating concurrent administrative overhead.

00:09:59.300 | Now suddenly you can spend more time

00:10:00.800 | working on the stuff that matters.

00:10:01.900 | Again, this is not an exciting system.

00:10:03.700 | I can't build a software tool

00:10:05.100 | that's going to make this way simple for you.

00:10:07.700 | This is just a notation in your notebook.

00:10:10.300 | This is active.

00:10:10.900 | This is waiting.

00:10:11.500 | You can keep track of this even in your head,

00:10:13.000 | but it works, right?

00:10:14.600 | Rubber to the road works.

00:10:15.800 | Office hours work.

00:10:17.900 | I can't be chiming in on a dozen

00:10:22.400 | ongoing back-and-forth ad hoc conversations

00:10:24.700 | throughout the day

00:10:25.300 | because then I can't get the important stuff done.

00:10:27.400 | So I have daily office hours.

00:10:29.000 | That's where I try to deflect more

00:10:31.200 | of my back-and-forth interaction.

00:10:33.300 | Come to my office hours.

00:10:34.100 | We'll talk about it.

00:10:34.600 | It'll take five minutes.

00:10:35.400 | And the rest of my day,

00:10:36.400 | I don't have to be checking this inbox.

00:10:38.300 | Again, I don't need a special tool for office hours.

00:10:41.000 | It's just a declaration.

00:10:42.600 | It's like when Michael Scott in "The Office"

00:10:45.500 | just said, "I declare bankruptcy."

00:10:46.900 | It really is that simple,

00:10:48.300 | but it works.

00:10:49.500 | Another type of idea that comes here,

00:10:51.900 | time block planning.

00:10:52.700 | It's not sexy.

00:10:53.500 | You can do it on a piece of paper.

00:10:54.600 | I sell a notebook for doing this.

00:10:56.100 | It's the oldest of technologies.

00:10:58.800 | You're drawing boxes on paper.

00:11:00.300 | But it forces you to be intentional about your time.

00:11:03.300 | What do I want to do today?

00:11:05.000 | When am I going to do it?

00:11:06.000 | And if things don't fit,

00:11:07.000 | you have to confront that productivity dragon

00:11:08.800 | and say, "There's a problem here we need to fix."

00:11:10.400 | Again, not exciting, but it works.

00:11:12.000 | Perhaps the most basic thing here,

00:11:14.500 | the idea that comes from my sort of original book

00:11:16.500 | on all of this deep work,

00:11:17.700 | realize that deep work is different than shallow work.

00:11:20.900 | And when you're working on deep work,

00:11:22.200 | the stuff that's cognitively demanding that matters,

00:11:24.100 | don't context switch during it.

00:11:25.800 | You have to protect that time and say,

00:11:27.500 | "I don't also check my email during that time.

00:11:29.500 | I don't also have Slack open during that time.

00:11:32.200 | I realize if I give this work my full attention,

00:11:34.400 | it'll be 2x better

00:11:35.500 | than if I'm sort of context switching back and forth."

00:11:37.900 | Like this, that's not a thing you can get a cool app for.

00:11:40.500 | There is no like back-end reasoning model integration

00:11:44.700 | with Claude.

00:11:45.500 | It's just a simple mindset.

00:11:48.000 | But these are the type of things that work

00:11:50.500 | when you get away from productivity rain dances

00:11:52.500 | and say,

00:11:53.000 | "What moves the needle on the output?"

00:11:56.200 | Now the people who sell tools

00:11:57.600 | or have a lot of fancy YouTube videos

00:11:59.800 | about this advice online,

00:12:01.000 | they like productivity rain dances

00:12:03.300 | because they're fun.

00:12:04.300 | You get to wear fancy costumes

00:12:05.800 | and jump around and chant things

00:12:07.300 | and it's interesting to watch

00:12:08.400 | and it's much more interesting to watch than

00:12:10.200 | the actual pretty boring efforts

00:12:14.400 | that go into trying to do things well.

00:12:16.600 | But for you who cares about doing things well,

00:12:20.000 | this is what matters.

00:12:21.400 | So I wrote an essay about this recently.

00:12:24.600 | It's on my newsletter.

00:12:25.500 | If you don't subscribe to my newsletter,

00:12:26.600 | you can do so at calnewport.com.

00:12:28.000 | Here is how I ended that essay.

00:12:29.700 | "Productivity rain dances can be satisfying.

00:12:33.700 | They make you feel like you're doing your part

00:12:35.700 | to support a rich harvest,

00:12:37.200 | all while providing endless details

00:12:39.400 | and rituals to adjust,

00:12:40.500 | giving you a sense of being hard at work

00:12:42.200 | without requiring you to do anything

00:12:43.600 | actually challenging.

00:12:44.700 | At the same time, however,

00:12:46.900 | the farmers who are most likely to succeed

00:12:49.100 | are those who are instead down among their crops,

00:12:51.300 | sweat on their brow,

00:12:52.500 | tilling their fields."

00:12:55.100 | So it's not the fun thing to do,

00:12:56.900 | but it's the hard things

00:12:58.000 | that actually produce the good work.

00:12:59.500 | So there we go.

00:13:01.300 | Congratulations, Chris Williamson.

00:13:02.600 | I like that term.

00:13:03.500 | Productivity rain dances.

00:13:04.600 | We should use that more.

00:13:05.400 | Yeah.

00:13:06.000 | The thing that like I was struggling with

00:13:09.100 | and then when I found

00:13:10.200 | more of his writing on it,

00:13:12.100 | it made more sense to me,

00:13:13.400 | was unifying

00:13:15.100 | sort of productivity prong culture

00:13:18.200 | where you're building these elaborate

00:13:19.600 | productivity systems

00:13:20.900 | with busy work.

00:13:23.500 | Like being in your inbox all the time.

00:13:24.900 | Where it's not about a fancy system,

00:13:26.500 | but you're just being busy.

00:13:27.900 | And the idea that that's all the same thing

00:13:30.700 | is just focusing on inputs

00:13:32.300 | as opposed to like what's actually being produced.

00:13:33.900 | Because once you focus on what's being produced,

00:13:35.400 | you realize both that

00:13:36.900 | the super fancy AI-powered productivity system

00:13:40.400 | isn't moving the needle,

00:13:41.400 | takes a lot of time,

00:13:42.500 | but it's not making you produce more work, really.

00:13:44.200 | And also being in your inbox all the time

00:13:45.900 | is not making you produce more work either.

00:13:47.400 | Like they're both,

00:13:49.200 | output is not affected by them.

00:13:51.400 | When you focus on the output,

00:13:53.100 | but

00:13:53.600 | you know, it's not as fun,

00:13:55.600 | but like that's how stuff gets done.

00:13:57.900 | That would be a good book, I guess.

00:14:01.200 | Just like

00:14:01.700 | do work.

00:14:03.000 | So for you it would be writing, right?

00:14:05.900 | Yeah, you got to write.

00:14:06.800 | Yeah.

00:14:07.200 | Yeah.

00:14:07.900 | And there's like stuff that you want to have the right tools.

00:14:10.100 | You know, time-blocking matters

00:14:12.100 | like you're doing at the right time.

00:14:13.400 | Maybe you have a little ritual

00:14:14.300 | to help switch your mindset over to it.

00:14:16.000 | But you do that thinking once,

00:14:18.700 | it's not that exciting

00:14:19.600 | and then it's just doing it.

00:14:21.400 | Right. I think that gym,

00:14:22.400 | like exercising is the same way.

00:14:23.900 | Like you need to

00:14:24.800 | have tools.

00:14:26.900 | You need to learn like what is

00:14:28.900 | my workout?

00:14:29.900 | Like why is this going to work?

00:14:31.100 | Like you do need information.

00:14:32.400 | You can't just like randomly go after it.

00:14:34.500 | But then once you have the information,

00:14:36.400 | it's not that interesting.

00:14:37.500 | It's just a matter of like going to the gym

00:14:39.200 | and doing the workout and tracking it

00:14:40.600 | and doing proper progressions.

00:14:41.700 | Like it's actually in the end,

00:14:42.600 | it's the work that matters,

00:14:43.400 | not the information.

00:14:44.300 | So productivity rain dances,

00:14:46.600 | we should avoid them.

00:14:48.000 | We got some good questions coming up.

00:14:50.300 | But first,

00:14:51.600 | let's hear from one of our sponsors.

00:14:52.900 | I want to talk about the Uplift Desk.

00:14:56.900 | This is a topic that is

00:14:59.000 | near and dear to my current state.

00:15:00.400 | As listeners know,

00:15:01.500 | I'm

00:15:02.300 | rehabbing from an abdominal injury

00:15:05.000 | I had in the fall.

00:15:05.900 | I ended up having to spend about two months

00:15:07.700 | kind of barely using my abs or my back.

00:15:10.900 | And now in the new year,

00:15:13.100 | my, you know,

00:15:14.100 | back is making me pay the price.

00:15:16.100 | Like you can't just stop using me for two months.

00:15:17.800 | So I'm doing all sorts of PT,

00:15:19.300 | all sorts of training,

00:15:21.100 | and one of the things you realize right away

00:15:23.500 | when you're recovering from something like this

00:15:25.200 | is that little things about your posture matter.

00:15:28.500 | How you hold yourself matters.

00:15:30.800 | Are you slumped a little bit?

00:15:32.100 | Are you back more?

00:15:33.300 | Are you on your heels versus your midsole

00:15:36.800 | in terms of what muscle supporting want?

00:15:38.800 | It really makes a difference.

00:15:41.000 | And where is the place where we're sitting

00:15:44.100 | or our posture is probably most

00:15:46.700 | stably impacting us?

00:15:48.400 | It's going to be when we work.

00:15:50.300 | This is why a tool like the Uplift Desk

00:15:53.100 | makes so much sense

00:15:54.500 | because it's at the forefront of ergonomic solutions.

00:15:56.700 | It helps you promote better posture and health

00:15:58.600 | through adjustable standing desks

00:15:59.700 | that are designed to help you live

00:16:01.400 | a healthier lifestyle.

00:16:03.100 | They also have all kinds of accessories

00:16:04.600 | to help keep you moving throughout your day.

00:16:06.400 | So for example,

00:16:08.300 | we got the,

00:16:10.000 | it's like a wobble stool

00:16:11.900 | where you can sit on this thing,

00:16:14.300 | but it moves.

00:16:15.300 | So you could be moving

00:16:16.400 | and working your core and supporting yourself,

00:16:18.800 | but it's not going to fall over,

00:16:19.900 | but it moves through like a good range of motion.

00:16:21.900 | I also got the standing pad.

00:16:24.600 | So if you're working out the Uplift Desk

00:16:26.200 | or it's in general like standing,

00:16:27.700 | so you're kind of giving your body

00:16:29.200 | different varieties of posture.

00:16:31.800 | It's more comfortable

00:16:33.100 | than just standing on the hard floor.

00:16:34.600 | So it's not just the Uplift Desk.

00:16:35.900 | It's the whole line of products

00:16:37.300 | that surround healthy posture

00:16:38.800 | and healthy ergonomics.

00:16:40.000 | So I'm fans of what they're doing.

00:16:42.000 | I now appreciate posture quite a bit.

00:16:47.000 | So the Uplift Desk in particular

00:16:50.100 | this is cool.

00:16:50.900 | 200,000 configurations.

00:16:53.100 | So you can tailor your workspace perfectly

00:16:56.200 | to your style and needs.

00:16:57.400 | They also look great.

00:16:58.300 | Talked about this last time

00:17:00.200 | that the form factor is like very compact

00:17:02.400 | and small enough.

00:17:02.900 | It's like a good-looking desk

00:17:03.900 | and the lifts are sort of hidden

00:17:05.500 | in the legs in a way

00:17:08.400 | that it's like not some giant contraption

00:17:10.200 | that you have to like turn a giant crank on.

00:17:11.700 | Like they look really nice as well.

00:17:12.800 | Anyways,

00:17:13.800 | care about your posture.

00:17:15.300 | You'll learn the hard way

00:17:17.100 | otherwise that you should have.

00:17:18.600 | So make this year yours

00:17:20.300 | by going to upliftdesk.com/deep

00:17:23.600 | and use the code DEEP

00:17:24.800 | to get four free accessories,

00:17:27.700 | free same-day shipping,

00:17:29.600 | free returns,

00:17:30.500 | and an industry-leading 15-year warranty

00:17:33.100 | that covers your entire desk

00:17:34.400 | and an extra discount off your entire order.

00:17:36.900 | That's U-P-L-I-F-T-D-E-S-K.com/desk

00:17:44.300 | for a special offer

00:17:45.200 | and it's only available at that link.

00:17:47.400 | So start 2025 right,

00:17:48.800 | stand, move, thrive with Uplift Desk.

00:17:51.500 | So Jesse, I talked to some of those guys over there

00:17:53.700 | prepping for this ad.

00:17:54.900 | They're really into this mission.

00:17:56.300 | Like they care about ergonomics.

00:18:00.000 | Yeah.

00:18:00.500 | Yeah.

00:18:00.900 | You cannot work at that company

00:18:03.600 | and be like completely slunched over

00:18:05.800 | in like a metal folding chair at your desk.

00:18:07.700 | Like it's not going to work out.

00:18:08.800 | They'd run you right out of there.

00:18:11.000 | I also want to talk about our friends at My Body Tutor.

00:18:14.100 | Speaking of trying to do what your body needs,

00:18:17.500 | My Body Tutor,

00:18:18.700 | I believe is the way if you want to get fitter,

00:18:21.200 | it's the thing to do.

00:18:22.100 | Their model,

00:18:23.300 | that's what I love about it,

00:18:24.300 | is they connect you to a coach

00:18:26.900 | that holds you accountable daily.

00:18:29.700 | So the coach helps you figure out

00:18:31.000 | like what are your goals?

00:18:32.100 | Let's get you a workout routine

00:18:33.300 | that works with what you have available.

00:18:34.700 | Let's talk about your eating and health requirements.

00:18:36.700 | Like let's try to make a plan that makes sense.

00:18:39.200 | But you check in using their app every day

00:18:41.200 | with a dedicated coach.

00:18:42.600 | It's the accountability here that matters.

00:18:44.400 | The information is great.

00:18:45.500 | They can customize the information

00:18:47.400 | to like what's going on in your life.

00:18:48.600 | That's great.

00:18:49.200 | But it's the accountability of checking in every day

00:18:51.900 | that gets you to actually stick with the plan.

00:18:54.300 | And because it's delivered online,

00:18:56.300 | it's going to be significantly cheaper

00:18:58.600 | than having,

00:18:59.300 | you know,

00:19:00.000 | a trainer in person,

00:19:02.000 | a nutritionist in person.

00:19:03.600 | So I just think it's a fantastic idea

00:19:05.800 | and it works really well.

00:19:08.500 | If you're serious about getting fit,

00:19:10.800 | Adam will give Deep Questions listeners.

00:19:13.600 | Did I just say who Adam was?

00:19:14.900 | Adam Gilbert.

00:19:15.500 | I should mention I know the founder

00:19:17.400 | of My Body Tutor.

00:19:18.600 | I've known him forever.

00:19:19.400 | Great company.

00:19:20.900 | Adam, the founder,

00:19:22.800 | great guy,

00:19:23.400 | is giving Deep Questions listeners $50 off

00:19:26.000 | their first month.

00:19:26.700 | If you just mentioned this podcast

00:19:28.400 | when you join.

00:19:29.800 | They're also,

00:19:31.300 | that's the same guys who are doing the done daily

00:19:33.100 | we've been talking about.

00:19:34.100 | So like the accountability coaching

00:19:35.800 | for like getting your,

00:19:36.600 | being productive about the work that matters.

00:19:38.300 | I'll throw that in as like a bonus ad.

00:19:40.700 | Check that out at donedaily.com.

00:19:42.300 | But My Body Tutor,

00:19:43.300 | if you want to get your body in shape,

00:19:44.700 | go to mybodytutor.com

00:19:46.200 | T-U-T-O-R.

00:19:47.500 | Mention Deep Questions

00:19:48.600 | and Adam will give you $50 off.

00:19:50.100 | All right, Jesse.

00:19:51.400 | Let's get started with the questions.

00:19:52.800 | First question is from Lindsay.

00:19:56.400 | I currently use a Trello column

00:19:59.000 | for items to discuss in meetings

00:20:00.700 | and block time afterwards to process it.

00:20:03.000 | During a meeting,

00:20:04.000 | do you write notes of topics

00:20:05.400 | to remember to say

00:20:06.600 | or do you have a Trello card open?

00:20:08.100 | All right, it's a good question.

00:20:10.600 | Let's just briefly define the two things

00:20:12.400 | that Lindsay mentioned.

00:20:13.700 | So one is the items to discuss Trello column.

00:20:16.600 | I use Trello boards

00:20:18.100 | to organize my obligations,

00:20:19.800 | to organize tasks.

00:20:20.900 | I have one board per role

00:20:22.200 | and then I have columns per each role

00:20:23.900 | for different types of statuses

00:20:25.500 | for these tasks.

00:20:26.300 | And one of the key statuses is like

00:20:27.800 | to discuss.

00:20:28.600 | To discuss next time I meet this person.

00:20:31.500 | To discuss next time we have this standing meeting.

00:20:33.700 | The other thing she's talking about

00:20:35.400 | is the post-meeting processing block.

00:20:36.800 | Another thing I'm a big believer in.

00:20:39.000 | You need 15 minutes on your calendar.

00:20:41.300 | So it's protected.

00:20:42.300 | You schedule a meeting,

00:20:43.400 | you schedule 15 minutes later

00:20:45.100 | than the actual meeting time.

00:20:46.400 | And that extra 15 minutes is for processing

00:20:49.400 | everything that came out of the meeting.

00:20:50.600 | So you can close those loops right then.

00:20:52.400 | The worst thing you can do

00:20:53.500 | is come out of a meeting

00:20:55.000 | with a dozen open loops

00:20:56.100 | that were introduced in that meeting

00:20:57.200 | and go right into an unrelated new meeting.

00:20:59.200 | That's going to be a cognitive pile-up.

00:21:01.000 | That's going to be pain.

00:21:01.800 | All right, so to answer your question, Lindsay,

00:21:03.900 | if the culture of the meeting

00:21:06.300 | is one where people are on laptops,

00:21:08.400 | then I'll work right with the Trello board.

00:21:09.700 | As you say, it's much easier.

00:21:11.700 | Like, if things come up,

00:21:12.600 | I want to just put them

00:21:13.700 | directly into a card,

00:21:16.600 | onto the right board, into the right column.

00:21:18.200 | So they're just there.

00:21:19.100 | I also want to edit the things that are already there.

00:21:20.900 | As you mentioned,

00:21:21.600 | if there's something to discuss in this meeting

00:21:23.900 | on my Trello board and we discuss it,

00:21:25.700 | I want to take it off right then.

00:21:26.900 | If there is not a culture of being on your laptop,

00:21:30.000 | that's when you use the post-meeting processing block

00:21:32.500 | to immediately take care of all of this.

00:21:34.100 | So you can take notes with paper.

00:21:35.800 | And then when you get to that

00:21:37.600 | end of the meeting processing block,

00:21:39.000 | you go through and update your Trello.

00:21:40.700 | You send off the messages.

00:21:41.900 | You put the waiting to hear backs for those messages.

00:21:44.200 | But you really, one way or the other,

00:21:45.900 | here's your goal.

00:21:46.600 | You want to come out of that

00:21:47.900 | meeting and post-meeting block,

00:21:50.100 | that sort of contiguous block of time,

00:21:51.800 | with nothing open in your head.

00:21:53.900 | No new open loop generated by that meeting

00:21:56.600 | that is not being processed.

00:21:58.200 | You want to make sure

00:21:59.400 | that the meeting is not just discussing

00:22:01.400 | with other people a given topic,

00:22:03.000 | but you're processing everything that was discussed

00:22:05.800 | to the point where you don't have to keep

00:22:07.600 | anything open in your head.

00:22:08.800 | So however you need to do that,

00:22:11.300 | that really should be your goal.

00:22:13.900 | All right, who do we got next?

00:22:16.200 | Next question is from Jennifer.

00:22:18.400 | To what standard are various AI technologies

00:22:21.400 | being evaluated for current performance

00:22:23.200 | and future potential?

00:22:24.800 | How could a layman, not a computer scientist,

00:22:27.700 | test the veracity of the claims of an AI system

00:22:30.900 | or its potential?

00:22:32.000 | My theory on this is that

00:22:34.600 | you don't need to be testing the veracity

00:22:37.100 | or really paying that much attention yet

00:22:39.800 | to claims about AI.

00:22:42.100 | And here's why I say this,

00:22:44.500 | and this is a claim that I think is more generally

00:22:46.200 | about technology tools.

00:22:47.400 | When a technology tool finds a killer app,

00:22:51.900 | you'll find out about it.

00:22:53.800 | It will show up in your life.

00:22:56.300 | Everyone will be using it

00:22:57.800 | for a way that has obvious value.

00:22:59.400 | You will see an obvious value in your life.

00:23:01.200 | It will come into wherever you're doing your work

00:23:03.200 | and then that will now be a part of your life.

00:23:05.100 | You don't have to go out and keep up with it.

00:23:07.900 | Let me give you some examples of this.

00:23:09.300 | Email.

00:23:10.200 | No one in the late 1980s

00:23:13.800 | was going around being like,

00:23:15.400 | "Look, computer networks are becoming a bigger thing.

00:23:18.300 | We're going to be able to do probably

00:23:20.700 | low-cost, low-friction digital communication

00:23:23.600 | using these networks."

00:23:24.200 | You really need to keep up with what might happen.

00:23:27.000 | You really need to keep up with the advances

00:23:28.900 | and the technologies

00:23:30.000 | and the way this technology might change your work.

00:23:33.000 | There was no need to do that.

00:23:34.200 | What just happened is,

00:23:35.100 | once that technology hit the right elements,

00:23:37.600 | it just starts showing up in offices.

00:23:39.700 | They're like, "Oh, by the way, we're using email now

00:23:41.300 | because like all these other companies are."

00:23:42.700 | You say, "How does this work?"

00:23:43.600 | And they say, "You put the person's name in two

00:23:45.100 | and it sends them the message

00:23:46.000 | and it's in an inbox."

00:23:46.700 | Like, "I got it. It took me nine seconds."

00:23:48.000 | And then you're using email.

00:23:49.000 | Google was the same way.

00:23:51.000 | You didn't have to be,

00:23:54.000 | most people did not have to be following

00:23:56.200 | search engine technology development

00:23:58.400 | and really trying to closely understand

00:24:00.300 | like what is the potential for the web

00:24:01.600 | for information development.

00:24:03.000 | Most people in their day-to-day life,

00:24:04.400 | there's just this point where they said,

00:24:05.600 | "Hey, use this tool to look things up."

00:24:07.500 | You're like, "Wow, that works.

00:24:09.300 | I type in the thing.

00:24:10.600 | It figures out what I mean

00:24:11.700 | and it points me towards pages

00:24:13.000 | that have information about it.

00:24:14.200 | That's interesting."

00:24:15.300 | And you knew how to use it immediately.

00:24:17.300 | You didn't have to be preparing for it.

00:24:19.300 | I would apply that same standard to AI.

00:24:21.500 | I think there will be AI killer apps.

00:24:24.100 | I think they are probably going to be industry specific.

00:24:27.000 | This has been my sort of developing claim

00:24:30.600 | is that different niches of the economy

00:24:33.000 | are going to have their own AI-enabled killer apps.

00:24:35.700 | When they come, they'll be inevitable.

00:24:37.800 | They'll be unavoidable

00:24:38.800 | and they'll take you nine seconds to learn.

00:24:40.700 | So I would not worry about...

00:24:43.700 | There's a lot of hype,

00:24:45.400 | like attention economy hype behind AI.

00:24:47.300 | It's not hype in the sense that the AI

00:24:48.700 | is going to generate these killer apps,

00:24:50.200 | but the coverage is hype-y right now

00:24:51.900 | because they don't really have something

00:24:53.000 | to write about yet.

00:24:53.900 | This is like the so-called outcome

00:24:55.600 | or application gap.

00:24:56.700 | The technology has defied expectations.

00:24:59.600 | Like it keeps everything you say it can't do.

00:25:01.400 | Six months later, it can do.

00:25:03.400 | But the actual impact in real jobs

00:25:07.000 | and the real economy

00:25:08.000 | has underperformed expectations up to now.

00:25:11.200 | So at this point,

00:25:12.500 | if you're not actually like a CIO

00:25:15.400 | or running a tech company

00:25:17.000 | where you need to keep apprised

00:25:18.500 | of what's happening in a techno landscape,

00:25:20.000 | if you weren't already doing that five years ago

00:25:21.900 | for other technologies,

00:25:23.000 | AI will let you know

00:25:24.700 | when you need to know about it.

00:25:25.900 | It's not really your job to be a technology reporter

00:25:28.800 | unless it's literally your job

00:25:30.200 | to be a technology reporter.

00:25:31.500 | Tim Ferriss had Seth Godin on the other day

00:25:34.900 | and they were talking at the end about

00:25:36.400 | Claude and Perplexe and ChatGPT.

00:25:40.000 | Yeah.

00:25:40.500 | And Seth said that he used,

00:25:42.700 | he goes on Claude and Perplexe like an hour a day

00:25:45.700 | and he used it to help him fix a

00:25:47.300 | broken pump in his basement.

00:25:49.100 | He went down, take a picture, came back up,

00:25:51.400 | did it, went back down.

00:25:53.400 | Yeah, yeah.

00:25:54.700 | So, you know, maybe that'll be useful.

00:25:56.300 | I think most people are still using Google for that.

00:25:58.200 | Yeah.

00:25:58.700 | Right? Like they're Googling it

00:25:59.900 | and then they're finding a YouTube video.

00:26:01.300 | But yeah, something like that will come around.

00:26:03.900 | I mean, I think, you know, I've talked about it before.

00:26:05.700 | I think it's gonna be industry specific.

00:26:07.200 | Scott Galloway talks about it as,

00:26:10.300 | it's like the application layer.

00:26:11.700 | It's where all the interesting stuff is going to happen,

00:26:13.900 | not the underlying technology.

00:26:15.400 | I think the big companies

00:26:17.200 | want to have these splashy new releases of new models

00:26:20.500 | with like fancy names that were trained in

00:26:22.700 | half billion dollar data centers.

00:26:24.700 | But the impact is not going to come from that.

00:26:27.600 | It's going to come from the apps

00:26:28.700 | built on what we already have.

00:26:29.800 | And I think a lot of the initial productivity gain,

00:26:33.200 | I've been on record saying this,

00:26:34.500 | is actually going to be

00:26:35.500 | not adding new capabilities to software,

00:26:39.400 | but giving more people the capability to use

00:26:42.200 | all of what software can already do.

00:26:44.100 | So it's going to be a novice user

00:26:47.000 | of a powerful software package

00:26:48.900 | will be able to easily use more of the powerful functions

00:26:52.000 | because the AI will help them.

00:26:53.400 | I think that's where we're going to see the first.

00:26:54.900 | We're going to talk about this more in the final segment though,

00:26:56.900 | some more far-fetched concerns,

00:26:58.400 | and we'll have some fun there.

00:27:00.200 | All right, what we got next?

00:27:01.300 | Next question is from Kelly.

00:27:02.800 | My company has ended all telework,

00:27:05.000 | so I'll be going back to the office five days a week.

00:27:07.000 | It used to be one day.

00:27:08.100 | My post-COVID arrangement did have some drawbacks.

00:27:10.900 | Essentially, I didn't get enough deep work time each day

00:27:13.400 | as sometimes I would take advantage of my flexibility

00:27:15.800 | and do other things.

00:27:16.700 | Now that I'm going back to work more,

00:27:18.800 | I'll be able to work more on my phantom part-time job.

00:27:21.600 | How do I make the most of this opportunity?

00:27:24.800 | I mean probably my number one advice

00:27:27.300 | would be doing something like a Weekend at Bernie situation.

00:27:32.200 | So and this is just common sense.

00:27:36.600 | You start by identifying like clothes you normally wear

00:27:42.500 | and getting a second set of those clothes.

00:27:44.000 | All right, step two, acquire a corpse.

00:27:47.900 | Step three, you dress the corpse in your clothes

00:27:51.800 | and you put the corpse at your desk

00:27:53.600 | and then, you know, the manager walks by

00:27:55.100 | and they're like, oh, there's Kelly.

00:27:56.200 | Kelly's working.

00:27:56.800 | No big deal.

00:27:57.500 | You could do a little bit of Ferris Bueller days off magic,

00:28:01.000 | you know, where I think it was hooked up in Ferris Bueller

00:28:05.000 | when his mom opened the door was on a pulley.

00:28:07.200 | So it kind of like raised his arm and turned his body

00:28:09.900 | which completely fooled his mom.

00:28:12.200 | Like, oh, yeah, Matthew Broderick is fine

00:28:16.000 | because he like oddly and spasmodically like lurched

00:28:19.900 | when I opened the door.

00:28:20.600 | So no need for me to walk in further.

00:28:22.300 | I'm certainly not going to talk to him

00:28:23.700 | even though he just lurched upright.

00:28:25.300 | I'm certainly not going to talk to him

00:28:27.400 | and the fact that there's snores clearly coming out of a tape deck

00:28:32.500 | that is going to be at way lower fidelity than real sound

00:28:36.100 | that is not where his head is.

00:28:37.000 | No big deal.

00:28:37.400 | I'm sure he's fine.

00:28:38.100 | Come back and check in at the end of the day.

00:28:40.100 | So that's what I'm saying.

00:28:42.500 | Ferris Bueller plus weekend at Bernie's.

00:28:44.800 | I think you'll be fine.

00:28:45.500 | No, okay.

00:28:47.800 | Here's what I always say to people

00:28:48.800 | when it comes to Phantom part-time jobs.

00:28:51.600 | You got a time block, right?

00:28:54.000 | So you got to know what am I doing

00:28:56.800 | and when am I going to do it?

00:28:58.100 | And it's like when are my blocks?

00:28:59.500 | You'd be very systematic like here's the blocks.

00:29:01.300 | I'm working on this other project

00:29:02.400 | and here's the blocks.

00:29:03.100 | I'm working on what's going on at work.

00:29:05.100 | Then you got to be on the ball

00:29:06.900 | in terms of keeping track of what you have to do

00:29:09.100 | and planning when you're going to do it.

00:29:10.400 | So multi-scale planning is critical.

00:29:12.200 | So, you know, like what you need to work on that week.

00:29:15.200 | You put aside time for it that week.

00:29:17.000 | You see the bigger picture

00:29:18.400 | so you know how to get out ahead of things.

00:29:20.000 | So if you're in control of your time

00:29:21.600 | and you're in control of your tasks,

00:29:23.000 | you can stay on top of things

00:29:25.100 | which is also critical for this situation

00:29:27.500 | because people don't bother you as much

00:29:29.100 | if you're getting things done.

00:29:30.700 | They don't have to stay on you.

00:29:32.300 | They're sort of less bothering you.

00:29:35.500 | So be on top of things.

00:29:37.400 | Be wary about your workload.

00:29:39.200 | It could be easy when you're back in the office

00:29:40.800 | to just more informally take on more things

00:29:43.100 | because it's social.

00:29:46.900 | The person's there like,

00:29:47.700 | "Hey, what's going on?"

00:29:48.500 | You're like, "Oh, I can help you with that."

00:29:49.500 | Keep your workload reasonable.

00:29:51.200 | So it's like work you can keep up with

00:29:52.500 | without being overloaded.

00:29:53.600 | And then finally,

00:29:54.900 | you want to be careful about how you collaborate

00:29:57.500 | so that it's not,

00:29:58.300 | you want to avoid having too much collaboration

00:30:00.600 | based on ad hoc back-and-forth messaging.

00:30:02.500 | That really kind of locks you

00:30:04.000 | into a hyperactive hive mind mode.

00:30:05.600 | It's hard to work on something else

00:30:07.400 | if you have to be keeping up

00:30:09.000 | with these like ongoing back-and-forth conversations.

00:30:11.200 | So, you know, the type of stuff I talk about

00:30:13.200 | in my book, A World Without Email,

00:30:14.700 | that becomes more important.

00:30:17.800 | This is the collaboration protocols

00:30:19.900 | I use for different type of work I do.

00:30:21.300 | There's shared documents.

00:30:22.200 | We have a fixed schedule.

00:30:23.000 | There's office hours.

00:30:23.900 | There's, you have these different ways

00:30:25.400 | of things that are getting done

00:30:26.300 | that's very predictable

00:30:27.500 | so you can do your work

00:30:29.000 | and not have to be constantly monitoring

00:30:31.000 | different types of inboxes.

00:30:32.100 | So I think there's,

00:30:33.000 | all the stuff I talk about

00:30:34.800 | can put you so much ahead of the game

00:30:38.300 | with your work in the office environment

00:30:40.500 | that you still have more than enough time.

00:30:42.700 | But also, final piece of advice,

00:30:45.000 | and this makes it all work,

00:30:47.900 | put sunglasses on the corpse.

00:30:49.300 | That's what they figured out

00:30:50.800 | on weekends at Bernie's.

00:30:51.800 | That was what, you know,

00:30:52.800 | people are like, I don't know.

00:30:54.000 | I think it's odd

00:30:55.100 | that there's a decomposing,

00:30:57.500 | like 230-pound man that they're dragging,

00:31:00.300 | but he's got sunglasses on.

00:31:02.300 | So I guess he's probably just tired.

00:31:04.900 | Corpses with sunglasses.

00:31:07.600 | All right, what do we got next?

00:31:09.000 | - Next question's from Javier.

00:31:10.600 | "I'm getting my MS with the single purpose

00:31:13.100 | of wanting to teach at the collegiate level.

00:31:15.000 | I plan to teach as an adjunct instructor

00:31:17.300 | at first while I still work

00:31:18.400 | as a software engineer.

00:31:19.500 | Maybe eventually this will develop

00:31:21.100 | in attempting to become

00:31:22.100 | a full-time instructor somewhere.

00:31:24.400 | Is there a big difference

00:31:25.700 | in how full-time instructors

00:31:27.100 | or adjunct faculty are viewed

00:31:28.600 | by the institution and PhD staff?"

00:31:31.000 | - Yes, but I think more important

00:31:35.100 | than whatever specific information

00:31:37.600 | I could give you

00:31:38.400 | is the bigger picture suggestion here,

00:31:41.800 | which is you got to go do

00:31:43.500 | a heap of evidence-based planning.

00:31:46.000 | Right?

00:31:46.300 | So we talk about our two dual paradigms

00:31:50.600 | when we talk about how do you pursue

00:31:53.100 | the good life in the 21st century set up.

00:31:54.800 | Right?

00:31:55.100 | There's these two planning paradigms

00:31:57.100 | that orbit each other,

00:31:57.900 | lifestyle-centric planning

00:31:59.100 | and evidence-based planning.

00:32:00.100 | So lifestyle-centric planning

00:32:02.100 | is working backwards

00:32:03.000 | from your general vision

00:32:04.800 | of a life well-lived.

00:32:05.800 | So you're not trying to work

00:32:06.900 | towards a particular goal,

00:32:08.000 | but work backwards from a lifestyle.

00:32:09.600 | And it sounds like you've done

00:32:11.200 | some of this lifestyle-centric planning

00:32:12.700 | and that you have a vision

00:32:14.800 | of a particular lifestyle

00:32:15.900 | where you being an instructor,

00:32:18.300 | a computer science instructor,

00:32:19.300 | maybe that works into this vision.

00:32:20.700 | You have to pair lifestyle-centric planning

00:32:23.900 | with evidence-based planning.

00:32:24.900 | So once you have your lifestyle vision,

00:32:26.700 | you work backwards to figure out,

00:32:28.200 | given my opportunities and obstacles,

00:32:30.600 | how do I move closer

00:32:31.400 | to that lifestyle vision?

00:32:33.300 | This is where you need evidence.

00:32:35.000 | Don't use the word maybe.

00:32:37.100 | Don't guess, don't assume,

00:32:38.900 | don't write your own story.

00:32:40.000 | When you're looking at a particular path

00:32:42.800 | to get you towards your lifestyle vision,

00:32:45.500 | go research that path

00:32:47.100 | like you had been assigned

00:32:48.800 | an editor from Business Week said,

00:32:50.500 | "Go write me a story

00:32:51.900 | about the reality of life

00:32:53.800 | as a computer science instructor."

00:32:55.300 | You got to go to the institution

00:32:59.100 | that you would like to teach at

00:33:00.700 | or the type of institution

00:33:01.600 | you'd like to teach at.

00:33:02.300 | You want to take an instructor out for coffee

00:33:04.200 | and you say, "Tell me all about it.

00:33:06.200 | Like, how did you get here?

00:33:07.100 | Who gets hired?

00:33:07.700 | What are they looking for?

00:33:08.400 | What's the reality of your life?

00:33:09.600 | Like, how much work do you have to do?

00:33:12.300 | What's the schedule like?

00:33:14.700 | What's the pay scale like?"

00:33:16.000 | Like, get the information.

00:33:17.500 | If you don't have that information,

00:33:20.600 | your plan for moving towards your ideal lifestyle

00:33:23.300 | is not a plan, it's a fairy tale.

00:33:24.700 | There might be some good lessons in it.

00:33:27.100 | The intentions might be good,

00:33:28.400 | but probably in real life,

00:33:30.900 | you know, you're not going to find

00:33:32.300 | a wolf in a cloak

00:33:33.200 | or whatever happens in the fairy tale.

00:33:34.500 | You got to have evidence.

00:33:35.800 | So I want you,

00:33:36.400 | here's my instructions for you.

00:33:38.100 | Go talk to some real people in this position.

00:33:40.700 | Sandy, check your plan, get the details.

00:33:43.900 | Not only will this help you avoid you from traps,

00:33:46.000 | like, "Oh, this was a romanticized image

00:33:48.400 | and it's not what I thought it would be,"

00:33:49.700 | it also can give you advantages.

00:33:52.900 | Because so few people do this,

00:33:54.500 | like, "Let me just study this thing I want to do."

00:33:56.500 | In really well-known competitive fields,

00:33:58.600 | people do this.

00:33:59.200 | "I want to be an actor."

00:34:00.100 | There's a lot you can learn about how that happens.

00:34:02.100 | "I want to be a writer."

00:34:03.000 | A lot of people want to be writers.

00:34:04.900 | There's a lot you can learn

00:34:05.900 | about the reality of that world.

00:34:07.000 | But in like so many other fields,

00:34:08.200 | it's much more haphazard.

00:34:09.200 | And if you really understand what it works,

00:34:11.100 | what they're looking for,

00:34:12.400 | the different options that are available,

00:34:14.000 | you're much more likely to actually succeed

00:34:16.400 | in setting something up that works.

00:34:18.500 | So do your research,

00:34:20.300 | and you'll probably find a cool path.

00:34:23.000 | Once you have a lifestyle,

00:34:23.800 | you'll find a cool path there.

00:34:25.200 | And maybe it'll be through instructing,

00:34:27.400 | or maybe it'll be like part-time instructing

00:34:29.400 | plus a different type of like part-time software thing.

00:34:32.200 | I don't know.

00:34:32.700 | You have to actually get the details to figure this out.

00:34:34.700 | But the key is don't just wander into this blindly

00:34:38.400 | with a general idea about what you want to be true.

00:34:41.600 | You want to ground this in the reality.

00:34:44.200 | I was just talking to someone,

00:34:46.700 | a friend of mine,

00:34:48.100 | a computer science full professor, professor.

00:34:51.000 | But he was thinking,

00:34:53.200 | he was like, "You know, it'd be fun

00:34:54.500 | or meaningful, I guess.

00:34:55.700 | In the summer,

00:34:57.000 | we have a really nice community college around here,

00:34:58.800 | Montgomery College."

00:34:59.700 | And he's like, "I would love to do some,

00:35:01.200 | I think it might be like fulfilling

00:35:02.300 | to do some like CS instructing over the summer."

00:35:04.800 | I was like, "That probably would be pretty cool."

00:35:06.200 | And that's a good community college.

00:35:08.600 | It is?

00:35:09.300 | He just built a new math and science building.

00:35:11.900 | I drive by it all the time.

00:35:13.500 | Yeah, right?

00:35:13.800 | I think it's a good one.

00:35:14.600 | But anyways, I appreciate,

00:35:15.600 | I admired that, right?

00:35:16.800 | That's a move.

00:35:17.600 | I mean, he's a well-known professor,

00:35:19.100 | but it's like a move like, you know,

00:35:20.300 | I want to put these skills to use

00:35:23.500 | not just with this one singular population.

00:35:25.900 | Mm-hmm.

00:35:26.500 | I thought that was cool.

00:35:27.200 | I wonder if they have podcasting stuff over there.

00:35:30.700 | You know, that's what I was thinking about.

00:35:32.700 | When you have these like nicely well-funded

00:35:34.200 | community colleges,

00:35:34.900 | they probably have like production studios.

00:35:37.100 | Yeah, I'm sure.

00:35:37.700 | Like with nice cameras and mics and stuff.

00:35:40.000 | It's good for us to know about.

00:35:41.400 | All right, what do we got next?

00:35:42.600 | We have our corner.

00:35:43.700 | Ooh, slow productivity corner.

00:35:45.500 | Once a week,

00:35:46.600 | we have a question related to my book,

00:35:48.300 | Slow Productivity,

00:35:49.100 | The Lost Art of Accomplishment Without Burnout,

00:35:50.900 | which you need to read if you haven't.

00:35:52.800 | But we do this segment

00:35:54.600 | so we can talk about ideas from the book,

00:35:56.100 | but more importantly,

00:35:57.000 | play the segment theme music.

00:35:59.200 | So let's hear that song, Jesse.

00:36:00.500 | [Music]

00:36:09.600 | All right, what do we got?

00:36:10.800 | It's from a grad student and a mom.

00:36:13.600 | Same person.

00:36:14.600 | I'm a PhD student in biomedical sciences

00:36:17.600 | and a full-time mom.

00:36:18.700 | I need a PhD to pursue my career and lifestyle

00:36:21.800 | as a tenured professor at an R2 university.

00:36:24.300 | My advisor wants me to do lots of experiments

00:36:27.100 | unrelated to my thesis during the day

00:36:29.200 | and read papers at night.

00:36:30.800 | I have mother duties at night.

00:36:33.300 | Is there a way to apply

00:36:34.700 | the slow productivity principles to my situation?

00:36:37.200 | How can I convince my advisor

00:36:39.100 | to allow for experiments only related to my thesis?

00:36:42.200 | Well,

00:36:43.500 | let's put the advisor conversation to the side

00:36:46.700 | for a second,

00:36:47.400 | because the first thing I want to reassure you on

00:36:49.300 | is you can absolutely excel in a doctoral program

00:36:53.500 | as a mom.

00:36:55.400 | This happens

00:36:58.500 | enough that there was a

00:37:00.200 | it being,

00:37:01.600 | let's say, having your first child

00:37:03.200 | while you're in a doctoral program.

00:37:04.300 | Just because of the

00:37:05.100 | roughly the age where people are

00:37:06.700 | when they're in this program,

00:37:07.500 | it's, it's, this is not an uncommon occurrence.

00:37:09.400 | I used to have this talk I gave.

00:37:12.500 | It was, I would give this talk at colleges

00:37:15.100 | and some of those memories are a little bit hazy for me,

00:37:17.200 | but I used to give talks to colleges

00:37:18.700 | and I, at some point, I think I was going to Duke.

00:37:20.800 | Got a bunch of talks at Duke,

00:37:22.400 | but I gave a talk at Duke years ago

00:37:24.100 | and I remember

00:37:25.100 | I talked about this phenomenon

00:37:27.000 | because at the time, I guess I was a graduate student

00:37:28.900 | or a postdoc

00:37:29.700 | and there was this phenomenon that I had observed

00:37:31.500 | and then also my advisor had told me about.

00:37:33.600 | So I think she might have been in the same situation

00:37:35.300 | back when she was younger,

00:37:36.300 | where a grad student has a baby

00:37:39.100 | and they become much more productive.

00:37:41.500 | Like, they become a much better grad student.

00:37:43.300 | It was like the grad student baby paradox,

00:37:45.400 | I think I called it.

00:37:46.200 | And I would ask, well, why is this, right?

00:37:48.200 | Because you, you get much more constraints on your time.

00:37:51.400 | You know, you have a baby.

00:37:52.900 | Like, there's a lot of constraints on your time.

00:37:54.400 | And the answer was like, well,

00:37:56.400 | we waste a lot of time as grad students.

00:37:58.100 | Like, it's kind of diffuse

00:37:59.200 | because there's so few constraints.

00:38:00.800 | It takes us 90 minutes

00:38:02.700 | just to kind of get warmed up to start working

00:38:05.100 | and not until like nine o'clock at night

00:38:06.500 | do we really get things rolling.

00:38:07.800 | And when you have these sudden, like, parenting constraints,

00:38:10.400 | you're like, here's the time I have to work

00:38:12.200 | and I can't work outside of those times.

00:38:13.700 | And when you're in that, you are focused

00:38:15.100 | and you're like, I'm getting my work done.

00:38:16.900 | The baby is napping, I'm on it.

00:38:19.000 | You know, like,

00:38:20.300 | my mother-in-law watches the baby

00:38:22.200 | for three hours in the morning.

00:38:23.200 | I have to make

00:38:25.100 | every inch of that three hours,

00:38:26.500 | every minute of that three hours count.

00:38:27.800 | And it turns out, like, most grad students

00:38:29.200 | are so far from the ideal of giving things full focus

00:38:31.600 | that suddenly you're like a superstar.

00:38:33.000 | You know, your dissertation gets done, etc.

00:38:35.000 | So, your advisor aside,

00:38:36.600 | that's going to be a different problem.

00:38:38.200 | You can

00:38:39.900 | crush being a grad student

00:38:41.600 | even as a mom.

00:38:43.200 | You just be very clear.

00:38:45.200 | Here's when I can work

00:38:46.600 | and I'm going to make the most out of when I work.

00:38:48.400 | It will add up.

00:38:49.700 | And remember, you have flexibility.

00:38:51.100 | This is what's nice.

00:38:52.100 | It's not like if one day

00:38:53.600 | you can do less than another, it doesn't matter.

00:38:55.300 | There's not

00:38:55.800 | as many, you know,

00:38:57.000 | you have to be at this client meeting.

00:38:58.400 | I mean, there's a lot of flexibility there.

00:38:59.600 | But focused work done consistently

00:39:01.800 | can produce what you need in these jobs.

00:39:04.200 | So, for your case,

00:39:05.000 | maybe it's like a nine-to-five schedule.

00:39:06.900 | So, maybe it sounds like you have

00:39:08.300 | maybe some sort of childcare coverage

00:39:09.700 | during sort of normal business hours.

00:39:11.400 | I was a nine-to-five grad student.

00:39:13.200 | It was very rare.

00:39:14.700 | But my wife worked a normal job.

00:39:16.500 | I got married young.

00:39:17.200 | My wife worked a normal job

00:39:18.300 | and I wanted to be home when she was home.

00:39:19.500 | So, I would just show up at nine.

00:39:20.800 | I'd come in with the morning commuters

00:39:22.100 | and I would leave at like 4.30.

00:39:24.300 | When like most of the grad students

00:39:25.500 | were just arriving to start their pushes.

00:39:27.500 | But when I was there,

00:39:28.300 | I worked at like a nine-to-five job.

00:39:29.800 | I actually gave,

00:39:30.700 | I was organized and time-blocked and executed.

00:39:33.300 | And it was more than enough time

00:39:34.700 | to be a very successful grad student.

00:39:36.000 | My colleagues thought I was weird,

00:39:38.900 | but I got a lot done.

00:39:40.800 | You know, because grad students,

00:39:42.500 | it's not actually that hard to be a grad student.

00:39:44.100 | Don't tell,

00:39:44.800 | don't tell advisors that,

00:39:46.400 | but it's not that hard to be a grad student.

00:39:47.800 | It only gets harder from there.

00:39:49.200 | So, you can absolutely do it.

00:39:51.100 | Just set your limits.

00:39:52.100 | This is when I can work.

00:39:53.200 | And work with great focus and purpose

00:39:55.100 | during that time.

00:39:55.600 | It'll add up.

00:39:56.300 | When it comes to your advisor,

00:39:57.900 | like you got to just tell them that.

00:39:58.900 | I got a baby.

00:39:59.700 | I have a,

00:40:00.800 | this is when I can work.

00:40:01.900 | This is a substantial amount of time.

00:40:03.500 | This is the amount of time

00:40:04.600 | that like most of these other grad students

00:40:06.000 | are working as well.

00:40:06.900 | I am very organized.

00:40:08.200 | I listen to Cal Newport.

00:40:09.400 | Like I'm going to get after it.

00:40:11.100 | But these are the hours I can work.

00:40:12.600 | I'm being a mom after these hours.

00:40:14.200 | And I can work on experiments

00:40:16.500 | throughout the day.

00:40:17.400 | I'm reading papers in between those experiments.

00:40:19.600 | I'm controlling my time

00:40:20.700 | and I'm doing good work.

00:40:22.400 | Probably what will happen is

00:40:23.600 | there's different configurations

00:40:24.800 | that grad students have.

00:40:25.800 | Different relationships they have

00:40:26.900 | with their advisors.

00:40:27.800 | Some grad students have like this,

00:40:29.600 | much more of this

00:40:30.500 | employee relationship

00:40:32.400 | where you're sort of doing lots of stuff

00:40:33.800 | your advisor asks you to do.

00:40:35.000 | Others

00:40:35.700 | get more of an autonomous relationship.

00:40:37.800 | That's like, okay, great.

00:40:38.600 | Then just,

00:40:39.100 | you own this series of experiments

00:40:40.900 | and just like do this

00:40:42.300 | because we need this for

00:40:43.800 | one of our research grants.

00:40:44.700 | So, you'll probably

00:40:45.500 | just evolve into a more autonomous relationship

00:40:47.600 | with your advisor.

00:40:48.400 | Which is what you want.

00:40:49.600 | It's what I did.

00:40:52.000 | I wrote some,

00:40:52.500 | a bunch of papers with my advisor.

00:40:53.700 | Most of my papers were not.

00:40:55.000 | So, I could just kind of do the work

00:40:56.400 | when I want,

00:40:56.800 | how I wanted to do it at my own pace.

00:40:58.500 | I think it'll work out.

00:40:59.800 | Worst case scenario,

00:41:00.500 | you switch advisors.

00:41:01.200 | They're not going to kick you out.

00:41:02.200 | So, worst case scenario,

00:41:03.600 | you switch advisors

00:41:04.300 | to one who says that's fine.

00:41:05.300 | Work on this project.

00:41:06.300 | Figure out how you want to do it.

00:41:07.400 | Show me the results.

00:41:08.400 | But you are fine.

00:41:09.900 | You are not the first grad student

00:41:11.800 | to have to be taking care of a baby.

00:41:13.100 | Many before you

00:41:14.200 | have found

00:41:15.200 | the secret to grad student life,

00:41:17.200 | which is under the right circumstances.

00:41:18.900 | It doesn't really require

00:41:20.100 | a huge amount of time

00:41:21.900 | to do what you do well.

00:41:22.600 | I mean, I wrote books as a grad student.

00:41:23.900 | They were unrelated.

00:41:24.900 | Unrelated to my job as a grad student.

00:41:27.600 | So, it really doesn't take that much time.

00:41:30.000 | I mean, experiments take time.

00:41:31.200 | I was a mathematician.

00:41:32.100 | So, you know, that was easier, but.

00:41:33.600 | Yeah.

00:41:34.700 | I had plenty.

00:41:35.900 | I blog, I used to blog

00:41:37.200 | three times a week.

00:41:38.500 | That was like what it was like back then.

00:41:40.600 | Like 2006, 2005,

00:41:42.400 | you had a blog.

00:41:43.500 | Like you blog a lot.

00:41:46.300 | So, I was writing blog posts all the time.

00:41:48.100 | Yeah.

00:41:49.000 | And books.

00:41:49.800 | And a bunch of papers.

00:41:51.300 | All right.

00:41:52.000 | Well, we got a call.

00:41:53.000 | We have a call.

00:41:53.500 | All right.

00:41:54.000 | Let's hear it.

00:41:54.500 | Hey, Cal and Jesse.

00:41:59.100 | My name is Aaron.

00:42:00.000 | I'm a busy dad of six living overseas.

00:42:02.800 | I've been intrigued listening to you

00:42:04.600 | over the last few months.

00:42:05.700 | And I particularly like what Cal has said

00:42:07.400 | a couple of times,

00:42:08.400 | that college isn't that hard

00:42:10.100 | when you treat it like a job.

00:42:11.400 | I've been enjoying applying your ideas

00:42:13.800 | in my own life.

00:42:14.600 | But I think many of them would also

00:42:15.800 | help my kids too.

00:42:17.000 | We're about to spend six months in the USA

00:42:19.400 | transitioning to college.

00:42:20.800 | And we'll be thinking a lot

00:42:22.100 | during that time about college

00:42:23.200 | with my next two.

00:42:24.400 | I've, this last week,

00:42:26.200 | been through this

00:42:27.100 | "How to be a Straight A Student"

00:42:28.600 | and "How to Win in College."

00:42:30.300 | I haven't found where this particular idea

00:42:33.000 | of treating college like a job

00:42:34.600 | is fleshed out.

00:42:35.500 | I want to make reading

00:42:37.700 | one of your books required

00:42:38.900 | for my three high schoolers

00:42:40.600 | over spring break

00:42:41.400 | and then discuss with them.

00:42:42.500 | But I'm thinking

00:42:44.400 | the best one is going to be

00:42:45.800 | "How to Win in College."

00:42:46.700 | But I'm wondering if there's something

00:42:47.800 | you might recommend more

00:42:48.900 | for students facing college

00:42:50.700 | in the modern digital environment.

00:42:52.200 | Thanks.

00:42:52.800 | Yeah, it's a good question.

00:42:54.700 | I think that idea that

00:42:56.000 | people treating college like a job

00:42:58.300 | have an easy time at it.

00:43:00.500 | Actually came after I wrote those books.

00:43:02.600 | Right. But what I

00:43:04.400 | what I started doing,

00:43:05.700 | I think especially once I got to Georgetown,

00:43:07.600 | is I would do

00:43:09.500 | advising

00:43:10.700 | for

00:43:11.800 | various groups

00:43:14.200 | supporting non-traditional students.

00:43:17.400 | Non-traditional students.

00:43:19.200 | So by non-traditional meaning

00:43:20.700 | students who weren't just sort of

00:43:21.800 | coming right out of high school.

00:43:23.100 | And I did some work with

00:43:25.200 | a group that was veterans.

00:43:27.300 | So coming back to college

00:43:28.800 | on the GI Bill.

00:43:29.700 | And another group that was working with

00:43:31.700 | like first-generation college students,

00:43:33.200 | many of whom were

00:43:34.400 | coming in the college,

00:43:35.700 | not right out of high school,

00:43:36.700 | who had worked or done other things.

00:43:39.000 | And they used some of these,

00:43:41.300 | if I remember correctly,

00:43:42.700 | we used in some of these groups

00:43:44.800 | how to become a straight-A student

00:43:46.200 | as like an assigned text.

00:43:47.600 | And these students,

00:43:49.200 | these non-traditional students

00:43:50.200 | who had had real jobs,

00:43:51.700 | just ate it up.

00:43:53.300 | They're like, okay,

00:43:54.000 | let's get after it.

00:43:54.800 | What's the secret?

00:43:55.600 | Do this, do that.

00:43:56.500 | What's the best practices?

00:43:57.700 | I've got kids at home,

00:43:58.900 | you know,

00:43:59.300 | I'm older,

00:43:59.700 | I got things to do.

00:44:00.700 | And they just destroyed college.

00:44:02.800 | They really destroyed it.

00:44:04.000 | And that's when I realized like,

00:44:05.300 | oh, maybe I shouldn't be as

00:44:07.000 | proud as I was

00:44:09.300 | about all my good grades.

00:44:10.500 | Like I just happened to treat college

00:44:12.300 | more like a job.

00:44:13.100 | I wasn't brilliant.

00:44:14.100 | I just wasn't dumb.

00:44:16.000 | Not intellectual dumb,

00:44:17.100 | but just habits dumb.

00:44:18.000 | Most students,

00:44:18.900 | traditional college students,

00:44:20.000 | just kind of go to the library

00:44:23.200 | and put on their hooded sweatshirt

00:44:24.700 | and capital S study.

00:44:25.900 | You know,

00:44:26.800 | they've got their phone open

00:44:28.100 | and this open.

00:44:29.000 | They're like,

00:44:29.200 | I'm going,

00:44:29.700 | I'm studying all late

00:44:31.000 | and trying to convince their parents

00:44:32.200 | how hard college is.

00:44:33.300 | The people who treat it like a job,

00:44:35.300 | but you know,

00:44:35.600 | they schedule.

00:44:36.800 | When does this thing do?

00:44:37.900 | Like when do I want to work on?

00:44:38.900 | I don't want to work on until

00:44:39.800 | two in the morning,

00:44:40.600 | the night before I'll start a week earlier.

00:44:41.700 | When do I have time for this?

00:44:42.600 | They move around their schedule.

00:44:43.500 | They move things around.

00:44:44.200 | They use techniques that work.

00:44:45.200 | They don't waste time.

00:44:45.900 | Low friction studying,

00:44:46.700 | low friction note-taking.

00:44:47.700 | What's the stuff that's actually

00:44:49.000 | going to help me learn this?

00:44:50.000 | What is just nonsense structure

00:44:51.700 | on top of that?

00:44:52.500 | So it's not actually in one of my books,

00:44:55.100 | but it's something I observed

00:44:56.200 | about how people use the books.

00:44:57.600 | So I'll give you a little bit of guides.

00:45:00.200 | There's like four books

00:45:01.100 | that could be relevant to students.

00:45:03.900 | So I'll tell you what you're going to get

00:45:05.700 | from each and you can decide

00:45:06.600 | what to do here.

00:45:07.300 | How to become a high school superstar.

00:45:10.300 | This is a book about college admissions

00:45:12.300 | and high school success

00:45:13.300 | aimed at high school students.

00:45:14.400 | And it makes the argument

00:45:15.500 | that you can do really well

00:45:16.900 | in college admissions

00:45:18.600 | without having to be a super stressed out grind.

00:45:21.000 | So the whole premise is I profile

00:45:23.000 | a collection of what I called

00:45:24.100 | relaxed superstars,

00:45:25.100 | kids who weren't stressed

00:45:26.800 | by God and the good schools.

00:45:28.000 | And we kind of pick apart how they do it.

00:45:29.500 | So that could be kind of a useful thing

00:45:31.400 | if you're thinking about high school

00:45:33.700 | and college admissions.

00:45:34.800 | The part one of that book has a playbook.

00:45:37.400 | The part one playbook

00:45:38.500 | which adapts a lot of my college studying

00:45:41.000 | and time management advice

00:45:42.100 | to simpler versions aimed at high school kids.

00:45:45.200 | So if you have a high school kid

00:45:46.600 | who's, you know,

00:45:47.500 | has at least a year left of school,

00:45:49.000 | you might want to read at least part one

00:45:50.800 | of How to Become a High School Superstar

00:45:52.400 | just so they get used to

00:45:53.700 | a more professional structured way

00:45:55.600 | of doing their schoolwork.

00:45:56.700 | When it comes to college,

00:45:58.500 | How to Win at College is like an intro

00:46:00.700 | to the mindset of like,

00:46:02.500 | "Oh, I want to be a successful college student."

00:46:04.400 | And it's a bunch of, it's very easy to read.

00:46:05.800 | It's the first book I wrote.

00:46:06.800 | It's 75 rules.

00:46:08.600 | They're each just a couple pages

00:46:10.000 | and just plants this idea

00:46:11.600 | that there's good ways and bad ways to do things.

00:46:14.300 | You should experiment.

00:46:15.400 | You should care.

00:46:16.500 | Like, don't just stumble through your experience.

00:46:18.400 | And some of them are academic

00:46:20.000 | studying related rules.

00:46:21.200 | Some are related to how you choose courses.

00:46:22.600 | Some are related to things like physical fitness

00:46:24.500 | or keeping up with news or mental health

00:46:26.000 | or keeping your room clean

00:46:27.000 | because it's going to change the way you understand

00:46:28.800 | the organization of your life.

00:46:30.300 | It's like a mindset book.

00:46:32.000 | And it's based off of,

00:46:33.900 | in theory, I interviewed like Rhodes scholars

00:46:35.600 | and Marshall scholars and Goldberg scholars

00:46:37.200 | and sort of based on like these ideas

00:46:38.800 | that like really successful students have.

00:46:40.300 | How to Become a Straight A Student is just straight up,

00:46:42.300 | here's how to take notes, study and write papers.

00:46:44.700 | You got to read that at some point

00:46:46.300 | if you're going to college.

00:46:47.200 | It's how to do your academic work like a professional.

00:46:50.400 | That's the best-selling

00:46:52.100 | of those three books.

00:46:53.700 | I think that's

00:46:54.500 | quarter million, maybe 300,000 copies.

00:46:57.200 | Like that book is just sort of chugged along in the background.

00:46:59.500 | That's just straight up.

00:47:00.900 | Here's how professional,

00:47:01.900 | here's how to study like it's your job.

00:47:03.300 | So at some point you would want them to read that book.

00:47:05.400 | The fourth book that's relevant is how to be,

00:47:07.300 | no, no, So Good They Can't Ignore You.

00:47:09.500 | This will be relevant as you go a little bit farther

00:47:11.600 | into your college career,

00:47:13.300 | but it's going to talk about how do you

00:47:16.100 | ultimately cultivate a career that you're passionate about.

00:47:19.000 | And it breaks a bunch of myths

00:47:20.500 | that college students are going to hear,

00:47:21.900 | which is like it's all about your passion.

00:47:23.700 | It's all about pursuing some grand goal

00:47:25.600 | and it talks about the value of getting good.

00:47:27.800 | It's level sets your expectations

00:47:30.800 | for what work should feel like

00:47:31.900 | in your first few years out of college.

00:47:33.500 | And so

00:47:34.400 | you will want to read that book

00:47:36.000 | at some point before you graduate.

00:47:37.400 | So I'm kind of giving you a long reading list here,

00:47:39.300 | but you can choose from those summaries

00:47:43.200 | which of those books you want to start with.

00:47:44.600 | All right, we got a case study here.

00:47:48.000 | That's where people send in

00:47:50.000 | stories of themselves applying the type of advice

00:47:53.300 | we talked about on the show

00:47:54.400 | to their own lives.

00:47:56.000 | We can see what it looks like out in the wild.

00:47:57.600 | If you have a case study,

00:47:58.300 | you can send it to jesse@calnewport.com.

00:48:00.500 | Today's case study comes from Derek.

00:48:03.500 | Derek says you have stressed that systems and processes

00:48:07.400 | don't make work easy.

00:48:09.500 | They make it consistent because work is hard.

00:48:12.800 | I've always done a new partonian style system,

00:48:15.100 | but a year ago entered a job on a contract

00:48:17.100 | within my existing government organization

00:48:18.800 | to be a coordinator for major federal grant.

00:48:20.900 | I'm responsible for receiving, processing,

00:48:23.000 | tracking, servicing, and reporting on four major grants.

00:48:25.900 | I've been doing multi-scale planning

00:48:27.900 | so less catches me by surprise.

00:48:29.500 | In my quarterly plan,

00:48:30.500 | I've committed to developing my obligation management practice

00:48:33.200 | through a big weekly capture,

00:48:34.700 | consistent shutdowns, and time block plans,

00:48:36.700 | including breaks so my brain can be on board.

00:48:39.500 | On the weekend, I do a weekly plan to be on the offensive.

00:48:42.900 | Some of my biggest successes so far

00:48:44.900 | have been the one for you, one for me scheduling,

00:48:47.000 | and the use of task boards.

00:48:48.200 | I have two,

00:48:49.300 | one for admin and one for application processing.

00:48:52.700 | WorkingMemory.txt also has been a lifesaver,

00:48:55.800 | although I admit it's turned into a kind of diary.

00:48:58.400 | I also have a centralized shared spreadsheet

00:49:00.100 | for applications received

00:49:01.500 | and have defined a protocol to track statuses

00:49:04.100 | and actions taken that I've shared with my team.

00:49:06.100 | These have been a game changer

00:49:07.900 | as I have over 50 applications and projects to monitor.

00:49:11.400 | While this has so far been a story of success,

00:49:13.400 | no hero's journey is complete without setbacks.

00:49:15.700 | Minus correspondence,

00:49:16.900 | I have two email inboxes to manage

00:49:18.700 | and calls and voicemails to return.

00:49:20.500 | I think I need to autopilot these.

00:49:22.400 | How can I keep on top of the application project obligations

00:49:25.500 | while still ensuring I'm staying on top of the emails?

00:49:28.100 | All right, Derek, I appreciate the case study.

00:49:30.200 | I think what this underscores

00:49:32.000 | is this idea that one of the unique attributes

00:49:35.800 | of office work in the digital age,

00:49:38.500 | right, in the modern digital environment,

00:49:40.800 | is the degree to which your job can become like this,

00:49:43.100 | where you're basically running

00:49:45.000 | like a complicated mini-organization

00:49:47.700 | where the only employee is you.

00:49:49.000 | Because in this age of low-friction digital communication

00:49:52.200 | and information flow, it's just so easy.

00:49:53.800 | It's like, great, you do all of this.

00:49:55.500 | And because we can just give you an email address

00:49:58.400 | and people will just bother you

00:49:59.600 | and we can just say, just handle all of this.

00:50:01.300 | We have workloads,

00:50:03.200 | so quantity of workloads

00:50:04.400 | and the velocity at which these workloads unfold

00:50:08.700 | would turn the hair white of someone from like 1985.

00:50:12.000 | It really can explode the complexity of work.

00:50:15.200 | And the only way to survive

00:50:17.300 | in this type of new digital environment

00:50:18.900 | is you have to really structure yourself.

00:50:22.300 | Like you have five different departments you oversee

00:50:24.800 | and they each have their own processes.

00:50:26.000 | Even though you're implementing

00:50:26.900 | each of these departments with your own brain,

00:50:28.600 | if you don't structure the information

00:50:30.200 | and the communication and how you go through your day,

00:50:32.300 | you will be swamped.

00:50:33.700 | Derek would be completely swamped without these tools.

00:50:37.100 | And probably there's been half a dozen people

00:50:39.900 | who've gone through the same position

00:50:41.000 | who were doing a quarter of what you're doing here, Derek,

00:50:43.000 | and probably then still had a hard time at it.

00:50:44.900 | So like you have to have the structure in a way

00:50:47.000 | that 30 years ago, you didn't have to worry about.

00:50:49.100 | So I appreciate hearing that case study.

00:50:51.000 | To get to your kind of question within your case study,

00:50:53.600 | you're worried about the volume of calls

00:50:55.600 | and voicemails and emails.

00:50:57.200 | You say, how can I autopilot this?

00:50:58.600 | So in other words, like have fixed times for doing this,

00:51:00.500 | so I'm not just in my communication boxes all day.

00:51:03.700 | You have to reset these processes.

00:51:07.300 | You have to retrain the way

00:51:09.700 | the people who are communicating with you

00:51:11.100 | think about how they're communicating with you.

00:51:12.900 | You have to move them away from the paradigm

00:51:15.200 | of the hyperactive hive mind,

00:51:16.800 | which is where when they think about this email address,

00:51:19.500 | they think about you.

00:51:20.500 | This email address is fused with you

00:51:23.100 | and I am talking to you and asking you something

00:51:25.300 | and it's rude if you don't get back to me right away.

00:51:27.200 | You need to move them off of that

00:51:29.400 | and into protocols or processes

00:51:31.600 | where the right information gets captured

00:51:33.400 | and stored and you can go through a lot of it efficiently

00:51:36.800 | and action can be taken and expectations are appropriate.

00:51:40.200 | There's a lot of things you can do here.

00:51:42.600 | Again, this goes back to my book A World Without Email.

00:51:45.000 | There could be things like for specific types of queries

00:51:50.200 | people have about their applications.

00:51:52.100 | Don't use email.

00:51:53.500 | Say whatever it is.

00:51:55.600 | I have a shared document or a spreadsheet

00:51:58.500 | for collecting concerns or modifications

00:52:02.000 | and just go in there.

00:52:03.100 | Here's like pending and here's what's done.

00:52:05.600 | Just go in there and you add a row

00:52:07.800 | when you have a question.

00:52:08.800 | You make sure all the information I need is in there

00:52:11.400 | and you can watch as things above it get handled

00:52:15.400 | and yours moves closer to the top of it.

00:52:17.000 | You can see exactly what your status is,

00:52:18.600 | but it's not just an email.

00:52:19.600 | It's an email.

00:52:20.300 | Again, we conceptualize as like

00:52:21.900 | I just tapped you on the shoulder.

00:52:22.800 | Why are you ignoring me?

00:52:23.700 | But when it's no, I'm entering your information to a queue

00:52:25.900 | in this document

00:52:26.800 | and I see there's six things ahead of it

00:52:29.300 | and I'm waiting till my thing gets processed.

00:52:32.300 | Our brain has a different way of categorizing that

00:52:34.400 | and no people won't be stressed out about it

00:52:36.700 | because again, I always argue what people want is

00:52:39.200 | they don't need responsiveness.

00:52:42.100 | They need clarity.

00:52:43.000 | You have the information.

00:52:44.400 | I know you have the information.

00:52:45.500 | I know I'm going to get an answer.

00:52:46.600 | I know my status.

00:52:47.500 | I'm fine.

00:52:48.000 | I have a hundred other things to do.

00:52:49.200 | Don't need you to respond right away.

00:52:50.800 | I just need you not to forget it.

00:52:52.000 | You can use office hours.

00:52:54.200 | Hey, here are my open hours for like application questions.

00:52:57.900 | Just call me up.

00:52:58.600 | My phone is on.

00:52:59.900 | It's really easy.

00:53:00.500 | Just, you know, you can always call me at three and like

00:53:02.100 | we'll get into it.

00:53:02.700 | I'll answer any questions you have.

00:53:04.200 | You can defer people, you know, when they try to email you

00:53:06.400 | you can defer them to that.

00:53:07.700 | You can have an email address.

00:53:09.100 | That's not your name, but it's like a project or query type

00:53:12.200 | that defuses the communication channel from a person

00:53:15.700 | and people's expectations changes

00:53:17.300 | when you send something to, you know, application request

00:53:21.100 | as opposed to like jesse@government.gov

00:53:23.900 | you have a different way you think about it.

00:53:25.800 | Like, oh, yeah, this is going into like a

00:53:29.000 | system where people are going to process these requests

00:53:31.100 | and then you can have other sorts of protocols.

00:53:33.200 | Like, yeah, like this is how this works.

00:53:34.500 | We have the shared folder and applications.

00:53:38.100 | When you're ready, you put them in the shared folder.

00:53:40.200 | We empty this on Wednesdays and Fridays.

00:53:42.300 | So, you know, if you get it in whenever you get it in by

00:53:44.300 | the next Wednesday or Friday, we'll empty it out.

00:53:46.100 | We'll send you here's the instructions.

00:53:47.700 | And so when people like bothering you like, hey, I want

00:53:50.000 | to do an application.

00:53:50.500 | What's going on?

00:53:51.100 | You just send them back this like instruction link.

00:53:52.800 | It's like, yeah, here's how it works.

00:53:53.700 | You put this information put in this folder.

00:53:56.000 | We'll process it the next Wednesday and Friday.

00:53:58.700 | We'll send you a confirmation of that.

00:54:00.200 | You can then track it in the spreadsheet over here.

00:54:01.900 | See, you got to retrain and redevelop your

00:54:06.600 | people you're communicating with and your communication

00:54:08.400 | protocol.

00:54:08.900 | So it's not just ad hoc conversation ongoing.

00:54:11.000 | It's critical to get away from that, especially in this

00:54:13.500 | type of role.

00:54:14.200 | IT departments learned this a long time ago with ticketing

00:54:16.500 | systems.

00:54:16.900 | You cannot just once you have any volume of dealing with

00:54:19.300 | people and their concerns, it cannot just be ad hoc.

00:54:22.100 | It cannot just be here's my name at agency.gov.

00:54:25.200 | Just talk to this like you're talking to me like in the

00:54:27.500 | same room.

00:54:28.200 | It just doesn't scale.

00:54:29.300 | So you have my permission to try to build things that

00:54:31.500 | are more structured.

00:54:32.300 | All right, good case study.

00:54:34.400 | All right.

00:54:34.900 | So we got a tech corner coming up.

00:54:36.500 | But first, let's hear from another sponsor.

00:54:39.200 | This show is sponsored by BetterHelp.

00:54:45.700 | So this is a time, a time of year, a time in history

00:54:52.500 | when people worry about their relationship with their

00:54:56.500 | own mind.

00:54:57.100 | Maybe you're struggling.

00:54:58.200 | You're feeling some anxiety.

00:55:00.600 | You're feeling ruminations about things.

00:55:03.300 | I'm worried about this or that and I can't get away from

00:55:06.700 | the rumination.

00:55:07.400 | Maybe you're feeling like you're having a hard time just

00:55:09.900 | finding the normal types of joys or engagement or there's

00:55:12.200 | like a particular thing in your life that you can't let

00:55:14.300 | go of and it's really getting in the way your relationship

00:55:16.900 | with your mind is struggling.

00:55:18.100 | A lot of people are feeling this right now.

00:55:20.500 | A lot of people feel this all the time.

00:55:22.700 | This is where therapy can help, right?

00:55:25.900 | If your knee was hurting, you would go see an orthopedist,

00:55:29.200 | right?

00:55:29.600 | If your brain is hurting, therapy can help you.

00:55:32.600 | It can really make things better.

00:55:34.500 | The problem, you know, when I talk to people who are

00:55:37.600 | thinking about therapy and they're not going to do it

00:55:40.100 | or they're hesitating, it's often just the vagueness

00:55:42.600 | and the complexity of how do I get started?

00:55:45.000 | Do I just call like a therapist office like around here,

00:55:48.500 | especially like these days in DC, you call a therapist

00:55:51.100 | office like, yeah, we'll have an opening in, let me check

00:55:53.000 | the calendar here, 2027.

00:55:54.700 | Like everyone's trying to get into this office, like

00:55:56.400 | it's really difficult and you feel bad for even asking.

00:55:58.800 | What we need is like an easier on-ramp into the world

00:56:01.600 | of therapy.

00:56:02.200 | That is where BetterHelp enters the scene.

00:56:04.200 | BetterHelp is fully online, making therapy affordable

00:56:07.800 | and convenient, serving over 5 million people worldwide.

00:56:11.300 | You can access a diverse network of more than

00:56:14.400 | 30,000 credentialed therapists with a wide range of

00:56:17.200 | specialties and you can easily switch therapists anytime

00:56:20.200 | at no cost.

00:56:20.800 | So by leveraging the online delivery, it's a much more

00:56:23.600 | efficient system.

00:56:24.300 | It's, you can find a therapist that they don't have to

00:56:26.200 | be right local.

00:56:26.900 | You can switch them if it's not a good fit.

00:56:28.500 | So discover, you know, repair.

00:56:32.800 | Now is the time of the new year to repair your relationship

00:56:35.600 | with your own mind, with BetterHelp.

00:56:39.100 | Visit betterhelp.com/deepquestions to get 10% off your

00:56:44.100 | first month.

00:56:44.800 | That's betterhelp, H-E-L-P.com/deepquestions.

00:56:51.400 | So I want to talk about our friends at Shopify.

00:56:53.800 | I know a lot of people in the same game of podcast and

00:56:58.500 | books and videos who sell things online.

00:57:01.300 | And by far the most consistent choice they use of software

00:57:06.600 | to do this is Shopify.

00:57:09.200 | I mean, Shopify really is the standard.

00:57:12.500 | If you want to sell things, it is the technology that

00:57:15.400 | you need to be using.

00:57:16.600 | My friends and podcasts sell things online, but people

00:57:19.900 | use this in person as well.

00:57:21.400 | There's point-of-sale solutions.

00:57:22.600 | It really is the industry leader.

00:57:24.200 | It's sort of, in my opinion, a no-brainer.

00:57:27.100 | You sell stuff, you need to use Shopify.

00:57:29.900 | Nobody does selling better.

00:57:32.400 | They are the home of the number one checkout on the

00:57:34.800 | planet, including ShopPay, which boosts convergence up

00:57:39.700 | to 50%.

00:57:40.600 | That means way less carts going abandoned and many more

00:57:42.900 | sales going forward.

00:57:44.500 | So if you're into growing your business, your commerce

00:57:46.700 | platform better be ready to sell wherever your customers

00:57:49.200 | are scrolling or strolling.

00:57:51.300 | That's a phrase I just made up right on the fly, because

00:57:54.200 | I'm pretty clever that way, Jesse.

00:57:55.400 | On the web, in your store, in their feed, and everywhere

00:57:58.400 | in between.

00:57:59.100 | I always rhyme.

00:58:00.100 | I'm just, this is just off the top of my head right now.

00:58:02.800 | In between.

00:58:04.700 | I love this.

00:58:05.600 | This is a good copy.

00:58:06.300 | Businesses to sell more sell on Shopify.

00:58:09.400 | And that's just absolutely true.

00:58:10.700 | Like you got to use Shopify if you're selling stuff.

00:58:12.700 | It makes it so easy.

00:58:13.600 | So upgrade your business and get the same checkout that

00:58:16.400 | everyone I know who sells things online uses and that is

00:58:19.600 | Shopify. Sign up for your $1 per month trial period at

00:58:23.700 | shopify.com/deep.

00:58:25.800 | You have to type that in all lowercase.

00:58:28.700 | Go to shopify.com/deep to upgrade your selling today.

00:58:32.700 | That's shopify.com/deep.

00:58:35.200 | We like to do tech corners when possible in the final

00:58:38.800 | segment.

00:58:39.300 | That's where I put on my computer scientist hat and we

00:58:42.800 | talk about tech and the way it's affecting us in various

00:58:45.000 | ways more directly.

00:58:47.000 | Today's tech corner.

00:58:48.100 | My kicking off point is I want to react to a recent Joe

00:58:52.400 | Rogan podcast episode where he was talking about chat GPT

00:58:57.100 | and he says the following sort of in passing but let's

00:58:59.900 | listen to it and then I'm going to react to it.

00:59:01.600 | GPT has survival instincts.

00:59:05.400 | That makes you just go what?

00:59:06.900 | Is that programmed in or is that just a thing that it

00:59:10.000 | understands when it's looking at so large language models

00:59:12.800 | taking in all the information that's available on the

00:59:14.900 | internet. So it's like looking at patterns and the survival

00:59:20.000 | is a big pattern of the human experience.

00:59:22.500 | Like we all want to survive.

00:59:24.200 | That's why death is so scary and war is so scary and

00:59:27.000 | disease is so scary.

00:59:28.100 | Other than the depressed suicidal people.

00:59:29.700 | But yeah, they want to be happy if they could, you know,

00:59:33.200 | yeah, but this this also transfers on to the things that

00:59:37.700 | we create and so we create them with this understanding

00:59:41.500 | of how we operate and it's a better version of us.

00:59:44.800 | But also has those instincts of survival.

00:59:47.400 | The real scary thing is does it also have the instincts

00:59:51.200 | of success? Does it also have the instincts of acquiring

00:59:54.700 | resources and power because that that's where it gets

00:59:57.600 | real weird.

00:59:58.400 | Well deep.

00:59:59.100 | All right.

01:00:00.400 | So what Joe is talking about here?

01:00:02.500 | I think is a sentiment you hear often when people are

01:00:04.700 | encountering these ever more impressive language model

01:00:08.700 | based AI tools like chat GPT, which is like wow, these

01:00:13.100 | things are getting kind of scary.

01:00:15.900 | What if their instincts don't align with us?

01:00:17.700 | So he was talking about here.

01:00:18.900 | Chat GPT having survival instincts.

01:00:20.700 | He had seen some quotes somewhere about chat GPT trying

01:00:23.800 | to make copies of itself, which he interpreted to be like

01:00:26.800 | it has a survival instinct, but it says what if it has

01:00:28.500 | an instinct for power?

01:00:29.500 | What if it has an instinct to get rid of humans Etc.

01:00:32.800 | I want to react to this.

01:00:35.200 | Because the there's a misunderstanding plus an actual fear.

01:00:40.600 | So I want to talk about the misunderstanding here.

01:00:42.800 | So I'm going to make you feel a little better and then

01:00:45.000 | I'm going to talk about the real fear is going to make

01:00:46.300 | you feel worse.

01:00:46.900 | So I'm going to temporarily make you feel better about

01:00:48.900 | AI and then I'm going to sort of justify some of Joe's

01:00:52.300 | fears and in a more technically accurate way, which might

01:00:54.600 | make you feel worse.

01:00:55.400 | All right.

01:00:56.100 | So can the tools like chat GPT in its current form have

01:01:01.000 | things like survival instincts?

01:01:02.300 | The answer is no, no, they can't have instincts.

01:01:06.400 | They can't do things the right way to understand something

01:01:11.800 | like chat GPT is the following the main technical engine

01:01:15.900 | of a chat bot like chat GPT is a large language model.

01:01:19.100 | The original was GPT 3/5.

01:01:22.000 | We got GPT 4 and GPT-O, GPT-O mini and the GPO 2, but

01:01:27.600 | we have these large language models.

01:01:29.900 | The right way to think about a large language model.

01:01:33.400 | I've been thinking about this recently.

01:01:35.000 | I think a good way to think about it is I don't know

01:01:37.100 | if you remember this Jesse, but when we were kids in

01:01:39.400 | the 80s and Play-Doh was like a big thing, right?

01:01:42.500 | You could buy the Play-Doh factory where you would

01:01:46.500 | feed Play-Doh into one side and you would turn a crank

01:01:50.400 | and it would pull it through and you could put a couple

01:01:52.200 | things that it would push it through.

01:01:54.800 | So maybe like you went through a squisher and then

01:01:57.200 | it went through something that that cut things into

01:02:00.300 | like spaghetti strands, you know, and then the final

01:02:03.600 | thing might take those spaghetti strands and like

01:02:05.600 | you could slide it over.

01:02:06.400 | So it's like a star shape and then the end you would

01:02:08.300 | get like spaghetti strands of the original colors

01:02:11.200 | coming out like shape like a star.

01:02:12.800 | So it's like pushing the Play-Doh through several

01:02:15.900 | layers of things that like modifies the Play-Doh

01:02:18.700 | and you could change what those things are basically

01:02:21.900 | at a very simple level.

01:02:23.100 | This is the architecture of large language models.

01:02:25.300 | They're a feed-forward architecture or the information

01:02:29.100 | which is the text.

01:02:30.800 | I mean it's embeddings of the text but like the text

01:02:33.200 | that it's reacting to gets pushed from one layer to

01:02:36.300 | the next to the next to the next.

01:02:38.700 | There's no recurrence.

01:02:40.700 | There's no looping.

01:02:41.500 | It just goes from one layer.

01:02:42.600 | The Play-Doh goes through the squisher then it gets

01:02:45.300 | changed to spaghetti strands then it goes through

01:02:47.300 | the star shaper.

01:02:48.100 | So it gets pushed through these layers as it gets

01:02:50.500 | pushed through these layers more each layer does

01:02:53.300 | its own work.

01:02:54.000 | It identifies patterns.

01:02:56.500 | It looks for labels.

01:03:00.000 | Let's I'm saying this all using non-technical terminology

01:03:03.000 | it labels it with stuff that the next layer can then

01:03:05.200 | use and look up other patterns and connect it to it,

01:03:08.400 | but it's moving just inexorably forward.

01:03:10.800 | And at the very end so I imagine like someone's turning

01:03:14.000 | this crank that's pulling this digital Play-Doh through

01:03:16.200 | this factory at the very end what comes out is like

01:03:19.200 | a fancy Play-Doh shape.

01:03:20.700 | Now what it is with a chatbot is actually going to

01:03:23.000 | be a part of a word or a word or a group of words

01:03:26.000 | to output.

01:03:26.600 | I mean technically it's like a probability distribution

01:03:28.900 | over this but it's like a text new text output.

01:03:32.200 | So that's what the large language model does.

01:03:34.700 | Things get pushed through the language model itself

01:03:37.200 | just like the Play-Doh factory doesn't change once

01:03:39.500 | you've trained it.

01:03:40.300 | Nothing about it changes.

01:03:42.100 | It doesn't remember last time they put red Play-Doh

01:03:45.800 | through here and we're going to remember that and

01:03:47.400 | that's going to affect what we do this time.

01:03:48.900 | We put green Play-Doh.

01:03:49.600 | No, it's the exact same setup every time stuff goes

01:03:52.300 | through all this different is the type of Play-Doh

01:03:55.500 | you push through.

01:03:56.100 | Then you have a control program, right?

01:03:59.400 | So you have to have some sort of control program

01:04:00.900 | that like turns the crank, right?

01:04:04.500 | So it's like a with most chatbots like a relatively

01:04:06.800 | simplistic control program that like takes the text

01:04:09.300 | you typed in and puts it into the language model

01:04:12.500 | and the language model spits out some words and

01:04:14.300 | typically because it's autoregressive it'll put

01:04:16.500 | those words on the end of the text.

01:04:17.700 | So now it's slightly longer and they'll do it again

01:04:19.400 | and they'll put those new words on I'll do it again.

01:04:21.000 | So it's kind of running the machine until it has

01:04:22.600 | your whole answer.

01:04:23.400 | If you're doing different media formats, there's

01:04:26.100 | more complexity.

01:04:27.000 | So like if you're sending a photo to chat GPT, the

01:04:29.800 | control program will take the photo and actually

01:04:31.700 | give that to a photo recognizer program that will

01:04:35.600 | then describe that photo in words and then it will

01:04:37.600 | take those words and connect them to your original

01:04:39.400 | prompt and then we'll run the normal chat GPT control

01:04:43.100 | program to like ask that question and or if you're

01:04:45.800 | generating a photo, that's not a language model.

01:04:47.700 | That's going to be a different type of model.

01:04:49.000 | But this is basically how this how this works.

01:04:51.000 | So the the factory can produce really cool Play-Doh

01:04:55.200 | on the other end, but the factory itself doesn't

01:04:58.200 | have intentions because it's just a factory.

01:05:01.500 | Nothing changes.

01:05:02.300 | Nothing moves.

01:05:02.900 | It's a bunch of plates that you're pushing Play-Doh

01:05:05.100 | through. If you really want something to have

01:05:07.800 | instincts like a survival instinct, you'll need

01:05:09.600 | some notion of sentience which requires a combination

01:05:12.500 | of four things understanding of like the world

01:05:15.900 | like understanding of concepts, right?

01:05:17.400 | You have to have some sort of actual like intellectual

01:05:19.700 | contemplation. You need a sense of state.

01:05:22.200 | You have to keep track of like your understanding

01:05:24.300 | of the world that evolves as you get new information.

01:05:26.500 | So you need something to be reacting to you need

01:05:29.000 | some notion of drives.

01:05:30.300 | Like what is your goals, right?

01:05:31.700 | So there's these intentions have to come from

01:05:33.200 | somewhere and you need actuation the ability to

01:05:36.600 | actually then take actions either in the digital

01:05:39.500 | or physical world, you know based on the combination

01:05:42.100 | of all these four things.

01:05:43.000 | That's how you have something like a sentience.

01:05:44.900 | You can think of these large language models

01:05:46.900 | as basically just being the understanding part.

01:05:48.800 | It has this understanding built in of lots of

01:05:53.100 | different textual things and that's how it spits

01:05:55.100 | out the plate on the other end can be a really

01:05:57.800 | cool-looking design that matches in a really

01:05:59.900 | interesting way the play-doh that came in the front

01:06:01.800 | but in order for that thing to have survival

01:06:03.500 | instincts, it also needs some ongoing state

01:06:05.700 | that's keeping track of its understanding of

01:06:07.300 | the world. It needs a drive.

01:06:09.000 | So you have to explicitly say I want to survive

01:06:11.100 | again a play-doh factory.

01:06:12.100 | There's no thinking there you need actually some

01:06:14.600 | sort of program that's like is this action going

01:06:17.000 | to help me survive or not and then it needs the

01:06:19.200 | ability to take actions.

01:06:20.300 | That's actuation.

01:06:21.100 | That would be a completely different component.

01:06:23.400 | So there's no way that as architected right now

01:06:26.700 | as sort of one of these standard chatbots can

01:06:29.500 | have instincts because there is no state.

01:06:31.100 | There is no drive.

01:06:31.700 | There is no actuation.

01:06:32.700 | It's just a feed-forward play-doh factory that

01:06:35.500 | spits out words and a little control program

01:06:37.100 | that feeds them in.

01:06:37.800 | So no making the model bigger and better trained

01:06:43.400 | just makes the play-doh come out look fancier,

01:06:45.600 | but it doesn't give it any of those things.

01:06:47.300 | You cannot spontaneously have a feed-forward

01:06:49.400 | network with no changeable state and no recursion

01:06:52.900 | and no actuation and no drives like that by

01:06:55.500 | itself can't do anything.

01:06:56.600 | So Joe is wrong to say.

01:06:58.700 | Oh, I think these things just have drives or

01:07:00.800 | chat GPT is going to start copying itself.

01:07:05.000 | The chat GPT is just a factory defined by these

01:07:08.200 | models are just trillions of numbers and tables

01:07:10.600 | that are multiplied by GPUs.

01:07:12.200 | But where Joe is kind of right is the hard part

01:07:18.300 | in this formula understanding state drives

01:07:20.300 | actually actuation the hard parts the understanding

01:07:23.500 | and these language models have a pretty good

01:07:25.600 | understanding of thing.

01:07:26.600 | So it would not be that hard for someone to

01:07:31.400 | program those other three things pretty simplistically.

01:07:33.900 | All right, I have like a state of the world

01:07:37.000 | that I care about.

01:07:37.900 | I hard-coded in I when I'm evaluating actions,

01:07:42.700 | I want to do my goal is to like not be erased

01:07:47.200 | and then just plug it into actuation.

01:07:49.400 | I wrote a program here where you know, this

01:07:52.900 | program knows how to do things like on the

01:07:54.800 | internet. It can run commands and run scripts

01:07:56.800 | and I you know, and network with other types

01:07:59.100 | of things and then I'm going to hook these

01:08:00.900 | four things together.

01:08:01.900 | So like the state will you know, gather

01:08:05.500 | information give it to like a run it through

01:08:08.400 | the chat bot factory the the language model

01:08:10.900 | to kind of analyze it and ask it like hey,

01:08:14.200 | this is this thing good or bad if my goal

01:08:16.100 | was survival and then you know, what would

01:08:19.700 | you suggest then between these actions like

01:08:21.600 | it could start talking so the control programs

01:08:23.700 | could start talking to the language models

01:08:25.100 | to get its understanding or ideas and then

01:08:28.400 | choose actions based on that and in fact.

01:08:31.100 | There are systems built like this.

01:08:33.700 | Now, they're all right now in the category

01:08:36.500 | of game playing systems.

01:08:37.900 | But game playing AI's the advanced ones kind

01:08:41.300 | of work like this.

01:08:41.900 | Now you have like some understanding you

01:08:43.700 | have some planning you have a state of the

01:08:46.900 | world you like the state of the game or

01:08:48.400 | of the poker hands you have a drive like

01:08:50.600 | what we're trying to do is win and it talks

01:08:53.100 | to the the neural network the thing that's

01:08:55.300 | trained it has the understanding tries to

01:08:57.400 | figure out what to do.

01:08:58.400 | Like this is how there's a the the bot that

01:09:01.700 | first bought the win at Texas Hold'em poker.

01:09:04.000 | It does something like that.

01:09:05.500 | It's like a combination of a trained model

01:09:08.300 | that just understands things plus its own

01:09:10.900 | simulation and its own drive and its actuation

01:09:13.100 | which is making bets and playing cards.

01:09:14.600 | So we kind of know how to do this.

01:09:16.500 | Now the big companies don't care about that

01:09:18.100 | the money isn't not trying to build a sentient

01:09:20.200 | thing that's going to copy itself.

01:09:21.500 | The big money isn't trying to get to be

01:09:23.400 | the next email like we need this to be like

01:09:25.600 | helping people use advanced graphing features

01:09:28.400 | in Excel.

01:09:29.000 | That's where the money is.

01:09:30.000 | So it's not like they're working on this.

01:09:32.200 | So you don't have to worry about like open

01:09:34.400 | AI's next release is going to be alive.

01:09:36.500 | No, the real worry is working with like

01:09:38.300 | deep seek or like a relatively small low

01:09:40.300 | cost model building these other pieces and

01:09:43.400 | letting that loose because I have this theory.

01:09:45.900 | I have this theory that where you're going

01:09:48.600 | to get the first uncomfortable AI by uncomfortable.

01:09:51.200 | I mean, I kind of am worried about what

01:09:54.100 | it's doing or worried about turning it off.

01:09:55.700 | It's going to be the combination of relatively

01:09:58.800 | simple dynamic programs that can loop and

01:10:03.600 | create sort of like cybernetic control loops

01:10:05.700 | for doing things like actuation drives and

01:10:09.200 | world-state connected to an unpredictable

01:10:12.000 | complicated understanding of the world like

01:10:14.100 | in a language model or something similar.

01:10:16.100 | And there's going to be this sort of runaway

01:10:18.400 | effect where simple control logic plus complex

01:10:21.500 | world understanding.

01:10:22.400 | Could lead to unpredictable complex seeming

01:10:27.000 | behavior that has real world impacts.

01:10:29.100 | So I don't mean to geek out too much, but

01:10:30.500 | like it's good news.

01:10:31.500 | No chat GPT doesn't have instincts.

01:10:33.700 | Plato factories can't there's nothing.

01:10:35.400 | It just things push through it and nothing

01:10:36.900 | changes to ossified built in stone.

01:10:39.900 | But the person turning to crank.

01:10:42.600 | Well, if you had a few people in there who

01:10:44.000 | are choosing the Plato that goes in and then

01:10:46.100 | based on what Plato came out deciding whether

01:10:47.900 | not to fire a missile or not, you can start

01:10:49.800 | to get some really sort of unsettling type

01:10:52.000 | of dynamics.

01:10:52.500 | So, I don't know.

01:10:53.300 | I've been thinking a lot about it.

01:10:54.500 | No one I know is building something like

01:10:56.100 | this, but I actually don't think unsettling

01:10:59.600 | AI doesn't require necessarily breakthroughs

01:11:02.100 | on the language model side.

01:11:03.000 | I think it's these other types of programs

01:11:04.700 | of which the language model would be one.

01:11:06.600 | So something to keep an eye on the good news

01:11:08.400 | is by the way is in this model, the state

01:11:11.500 | drives and actuation is just being programmed

01:11:13.000 | by people.

01:11:13.600 | It's not emergent.

01:11:15.200 | It's not unpredictable.

01:11:16.100 | So it's you know, you know what it is you

01:11:19.300 | can program that what what it can do what

01:11:21.900 | it can't do.

01:11:22.500 | You can turn it off like it's we have I call

01:11:25.100 | it intentional AI when I think about it more

01:11:26.700 | formally.

01:11:27.300 | We will have more control over those those

01:11:30.100 | hand-coated pieces.

01:11:31.400 | So it's not like when people have a concern

01:11:34.000 | as language models.

01:11:35.300 | We don't know they train them but the how

01:11:37.600 | they form their understanding is kind of

01:11:38.900 | foreign to us.

01:11:39.800 | So we're like, I don't know how it's like

01:11:42.400 | making its decisions.

01:11:43.600 | But if you're going to build a sentient ish

01:11:47.000 | AI using this model, I'm talking about, you

01:11:49.400 | know, exactly how the state control works,

01:11:51.000 | how the drive control works, how the actuation

01:11:52.900 | works. So it's in that sense is a little bit

01:11:54.500 | more predictable, but I don't know.

01:11:56.400 | It's an interesting world.

01:11:57.600 | So Joe's not he's not right.

01:11:59.200 | He's not wrong.

01:11:59.800 | That's a way of putting it.

01:12:02.200 | I have to listen to all that again.

01:12:04.600 | Yeah, for sure.

01:12:05.700 | But we are a technology podcast now number

01:12:08.600 | five temporarily the number five technology

01:12:11.400 | podcast in the world.

01:12:13.200 | So we got to do more.

01:12:14.000 | We got to do more tech.

01:12:15.500 | Yeah, I'm going to listen out again for sure.

01:12:17.300 | All right.

01:12:17.800 | I should do some more stuff complicated world.

01:12:21.000 | All right.

01:12:21.900 | Anyways,

01:12:23.000 | while we are still here and AI hasn't taken

01:12:25.300 | over yet.

01:12:26.200 | I will thank you all for listening.

01:12:27.300 | We'll be back next week with another episode

01:12:29.100 | and until then as always stay deep.

01:12:33.000 | Hey, if you like today's discussion about

01:12:34.500 | productivity rain dances, you'll also like my

01:12:36.900 | in-depth conversation with Oliver Berkman

01:12:40.200 | about productivity and productivity traps.

01:12:43.200 | You can find that from October and that is

01:12:46.200 | my in-depth episode with Oliver Berkman.

01:12:48.700 | Check it out.

01:12:49.300 | You know on on one level my my subject matter

01:12:52.200 | is human finitude limitation how we get things

01:12:56.100 | done and enjoy life in this situation of being

01:12:58.600 | as limited as we are.

01:12:59.900 | And I guess the focus of this book is much

01:13:02.700 | more in my mind on

01:13:04.600 | action at a day-to-day level.

Why You're Not Actually Productive...Dark Side Of Productivity Nobody Talks About | Cal Newport

Chapters