What's up, everybody? This is Cortland from IndieHackers.com, and you're listening to
the IndieHackers podcast. More people than ever are building cool stuff online and making
a lot of money in the process. And on this show, I sit down with these IndieHackers to
discuss the ideas, the opportunities, and the strategies they're taking advantage of
so the rest of us can do the same. I'm here with my brother, Channing Allen. Channing,
how's it going?
It's going swell.
This is your first time on the IndieHackers podcast. You've been my twin brother for
35 years. People don't know you because I don't talk about you on the show, but you've
been helping me run IndieHackers for the last five years, sort of my co-founder for IndieHackers.
You joined IndieHackers seven months after I started it, and then a month later, we got
acquired by Stripe. Patrick asked if I wanted anything as part of the deal, and I was like,
yeah, you got to hire my brother, too. And so you've been working on IndieHackers with
me ever since, but you've never once come on the podcast. I don't know why that is.
Have you never wanted to come on? I think I've invited you on.
Yeah, I don't know. I feel like we have a division of responsibilities we kind of always
have. And I honestly just, in my mind, I just partition what I was working on and a little
bit away from what you were working on. I don't even listen to half the episodes.
And what do you do? What do you work on?
Well, so we both do strategy, but you largely do the code stuff. I vaguely say you're sort
of more focused on the community historically, and I've been a little bit more focused on
the media side, except for, ironically, the podcast itself.
Okay, so yeah, and now you're finally on the podcast. And I think the impetus for this
is the fact that we were both just in Italy together. I don't really see you that often.
You live in New York City, you live in Seattle, but we both went to Bologna to basically hang
out with a bunch of other indie hackers at this founder conference thing that happened.
And it was pretty cool. I mean, I had a good time. I talked to Peldy, probably the funniest
guest I've had on the podcast. Tara Reade was there, Laura Roto was there, Rand Fishkin
was there, a bunch of different founders I've had on the podcast before. And I think that
was probably your first time meeting most of those people, right? What'd you think?
I really liked it. Part of that is also we were together, right, for a lot of the time.
And we literally just put a lot more time into dog fooding our website, like actually
going in and using it. And we just had way more opportunity to just both be sitting at
the kitchen table together, using the site.
If we're trapped in a room together, what else are we going to talk about? We're going
to talk about indie hackers, right? I don't want to hear about your novels for the 15,000th
time. You don't want to hear about my polyamorous lifestyle for the 100,000th time. But like
indie hackers, it's the thing that we both do for a living. It's endlessly entertaining.
So obviously we're just going to talk about that and then get all riled up over it.
But the downside was getting COVID. I have so far throughout the pandemic for two years
now managed to not get COVID that I know of. And I've been tested at least a dozen times.
Every time I feel the slightest bit sick, I get tested. And I have not tested positive.
We go to the conference, I don't know, two days afterwards, half the people there have
COVID. It's basically the super spreader event. And I finally go from being one of the cool
kids who's not gotten COVID in the last two years to one of the COVID kids. I feel like
a zombie in the zombie movie now, I got got. And you got COVID too. So we had to basically
be quarantined in Italy together for two weeks, much longer than I wanted to be in Italy.
There's worse places to be quarantined, but still, you don't want to be in a foreign country
sick with COVID. Ups and downs of COVID for you. And it was also your first time getting
COVID. What was the high point? What was the low point?
It was all low, but I'm actually more curious. I want to I want to like do a who's got the
better brother immune system test. How are you doing now? How do you feel now?
Oh, I feel good now. I mean, it's been like two weeks, it's been two weeks since we tested
positive. So I feel good. I feel totally healthy, except for the fact that like I've been sleeping
like 12 hours a day, 1012 hours a day, I normally sleep like five hours a day. And not because
I'm like some sort of like braggy, I don't need to sleep person. But because like, I
go to sleep the second my head, it's the pillow and four or five hours later, I'm just up
and I can't get back to sleep no matter what. And I'm just ready to start the day. And so
it's been like a total mindfuck for me to basically stay asleep for eight hours and
then wake up and then the middle of the day take another like two or three hour nap.
And I thought it was jet lag, because Italy is nine hours ahead of Pacific time. So it's
like, essentially, by the time like I'm going to sleep here on the west coast, like I would
have been waking up in Italy. So maybe it's partly jet lag. But I think partly it's just
COVID, right? It's like recovery from COVID. My body's just tired. So other than that,
I feel like a million bucks. Where are you at? I'm good. But I feel like I kind of waver
in and out. I mean, you were there, we got COVID, like right after we tested positive
for it, we began to actually have our symptoms a little bit. And then I played what was it
for three hours straight of VR ping pong. And then instantly, I stopped feeling good
for like two days. I perhaps some topics just to talk about that I thought would be interesting.
I think you did the same. Maybe you want to toss one out. So have you heard anything about
New York Times and their editorial board telling their reporters to spend a lot less time on
Twitter? Yeah, so this is this is super fascinating to me, because every major news organization
for the last like five years has really aggressively taken to Twitter, right? Every New York Times
reporter is on Twitter, they break their stories on Twitter, it sort of seems like they have
to. And it turns out New York Times, like internally, they almost had a mandate, it
wasn't official. But like, if you're a reporter, you're expected to be on Twitter. And that's
really quickly unwinding right now. There was a memo that went out, where their executive
editor basically said to turn that down. And I thought that was really interesting, because
even though the reasons they give for this are super super milk toast, like in the memo
is like, you know, hey, you know, there's sort of a risk of harassment on Twitter. You
know, when you're on Twitter, people are a little bit more distracted than I found a
summary, a summary of their memo, it's like four points for why they think their journalists
should spend less time on Twitter. Number one, Twitter takes up too much of the journalist
time. Number two, it warps their reporting by changing who they see as their audience
and the feedback to get on their work. Now they're just kind of catering to the Twitter
audience as if that's everybody. Number three, it's a major driver of harassment and abuse.
So number four, bad tweets are a significant reputational threat to the New York Times
and its staffers. So upshot of all that, get off Twitter, or stop using so much.
Exactly, exactly, exactly. So those are all the explicit reasons that were given. I heard
Cal Newport, the author of Deep Work has sort of has a podcast, they talked about a real
reason or what he believes is a real reason for this, which is a lot juicier and a lot
more interesting, which is one of the most popular writers at the New York Times, Taylor
Lorenz, you know her, right? She's constantly in these dust ups and gets into a lot of drama
where she like talks shit to Silicon Valley VCs and she tries to infiltrate what's going
on. She is now, I don't know if she got fired or she left the New York Times, but she basically
has gotten in these like arguments with people from the New York Times. So like Maggie Haberman,
she's like this really, she's like a White House reporter. She had all these scuffles
with like Donald Trump and Maggie Haberman and Taylor Lorenz have had like these like
public battles where they're arguing against each other and they're kind of talking crap
about each other in the news. And the two sides are sort of like this. Taylor Lorenz
basically says that the New York Times in a lot of ways suppresses, you know, sort of
her need to try to grow her personal brand. She can get paid a lot of money on sub stack
or she just got a really probably much more profitable high paying job at the Washington
Post. A lot of New York Times authors have left the New York Times and have started sub
stacks where they're making tons of money and it's because they're building their brands.
They're scuffling on Twitter. They're like getting embroiled in these dramatic fights.
And the tension is when they're doing that, they're not necessarily doing things that
are the best for the New York Times. I mean, every time Taylor Lorenz got in some kind
of a, got into an argument with people in the Silicon Valley VC arena or got in arguments
with like the tech world, she complained that the New York Times didn't back her up and
didn't like come and like support her. And from their perspective, they're like, well,
that's not our job, right? And this isn't your job. And so it's kind of funny because
this to me, we obviously talk all the time about personal branding and like right now
at the top of our forum, there's like a guy who's complaining that there are way too many
indie hackers who try to build their personal brands and spend too much time on Twitter
as opposed to focusing on their customers.
Well, I think that's what the internet is about, right? The internet is about authenticity.
It's about the fact that you can for the first time in history, just like, oh, you like an
author or a scientist or a politician, just tweet them and they can open it and they don't
necessarily have to have someone employed to do this for them and read your tweet and
your message and respond directly to you. And I think that part of what's happening
is that like, you know, the internet's been around for 30, 40 years or whatever the web
has. Part of what's happening is the culture is still changing around that just because
we've all adopted the technology doesn't mean our culture and our institutions and our businesses
have adopted too.
And I think one of the changes is that like, if you're the New York Times, you just kind
of have to eventually suck it up and realize that like, your reporters do kind of have
all the power. And even if you want to retain all the power in your organization, like your
reporters and journalists are the ones writing the stories. They're the ones talking on the
podcast. They're the ones talking on YouTube. Audiences are going to have more affinity
for them than they do for your brand or, you know, at the very least an equal amount. And
so they're going to build up followings. They're going to be able to command higher salaries
and to request higher levels of support from the parent brand. And if you don't keep them
happy, they'll leave.
And I think that that's just where the world is going. And so like people like Taylor Lorenz,
like, I think she's ultimately right. You know, she like left the New York Times go
to the Washington Post because they get the internet a little bit more than New York Times,
she says. Meanwhile, the New York Times is telling its reporters to try to like, you
know, clamp down on Twitter. I don't think they necessarily have the clout or the power
to do that, right? Like organizations that thrive will be the ones that cater to the
power of the individual journalist. There is some room for like, okay, aggregating all
of these people like the New York Times under one brand, like the New York Times does have
its own distinct brand, I think that will endure. But I don't I just don't think it's
going to retain as much power over individual creators as it has in the past. So instead
of trying to like play this losing game where we stick to the past, let's just play a winning
game where we try to like basically out innovate ourselves. It's really painful. It sucks,
right? But like you might as well do it. So like, okay, give someone like Taylor Lorenz,
give like your star journalists an amazing amount of upside, like allow them to become
millionaires by working for you do everything in your power to throw followers at them.
Right? If you do that, then like, why would they ever quit? Why would they ever leave?
Why would they ever go to a sub stack or something else? So another one I wanted to talk to you
about, I think it'd be interesting to you as this as this technology, it's this new
basically, I don't know how to refer, is it an algorithm? Is it? I don't know what it
is. It's called Dolly to an AI system that can create realistic images and art from a
description written in natural language. Have you seen this? Some blowing up on Twitter
the last couple weeks. I have. I haven't seen much. I've seen some, some of me to post.
It's pretty, it's pretty thick. And so there's like, do you know what GPT three is the big
sort of AI splash a couple years ago, a year or two ago, it's like, you can basically generate
text that looks like it's written by a human. And Dolly is the same thing, but it generates
images that look like they're created by a real artist. And like you can give it the
like most brief of descriptions, and it will spit out like an amazing image that looks
like it was created. Like you can say, for example, draw me a photo of a quaint flower
shop storefront with a pastel green and clean white facade and open door in a big window.
And it'll literally spit out like a photo of that kind of flower shop at a cool angle.
And it just looks like a photo that someone took of a real flower shop somewhere. It's
indistinguishable from a photo someone took somewhere. Or you can say, Oh, I want it like,
I want that in the style of like art deco. And it'll give you like an art deco painting.
That sounds like everyone's logo should be made by this. Exactly. It's fucking dope,
dude. It's like almost any art style you can imagine. This thing can within seconds spit
out a picture in that art style. Like I'm looking at some of these pictures, so many
people who have access to this are on Twitter, and their minds are blown and they're creating
threads of all these pictures. Like one of them is like an astronaut surfing on a sea
turtle over a rainbow past planets in space. And it's literally a picture of an astronaut.
It's like it looks like somebody painted this. It's like super colorful and artistic. And
I think to me, like the obvious implication is exactly what you're saying, which is like,
we no longer need illustrators. Like if I want to create a blog post, I might in the
past have spent a bunch of money to hire an artist in house. But now I can like, okay,
I can log into Dolly to give it, you know, some input to tell it about my style, keep
clicking refresh until it gets the style that I want. And then just describe every blog
post I write every article I write, and it'll just like create its own unique art to go
at the top of it that looks amazing. And why would I pay hundreds or thousands of dollars
for an illustrator if this computer can do it? And seconds, much faster and basically
for free.
I hope I hope our illustrator for our newsletter doesn't listen to this podcast. But that but
okay, but so that that it seems like when you want an illustration done, or when you
want almost any kind of art done, especially like at a commercial level, like, hey, this
company, you know, we do blog posts, or we do a newsletter, we do XYZ, like you want
it to seem fairly consistent. So I like that's kind of one of the only question that I have,
like is this thing to let me send you let me send you some of the pictures of this.
Like here's my favorite one. I just found I'm just looking at people's Twitter threads
that they're doing here. I just sent it to you on telegram. This one he typed in a bowl
of soup that's a portal to another dimension as digital art. And it gave him two different
versions of this art. One of them is like this top down view of this soup with this
cool rainbow swirl. That's like magical coming out of the soup. And it's like a painting
of that. And the other one is kind of a side view of this like blue soup with this other
like, mystical portal. It's just crazy. Here's the thing. People have said for years, that
computers cannot be creative. No matter what computers can do, you know, they're just dumb
calculators, they can't be and this is mostly coming from like, artists, right? The computers
will never replace people that can't be creative. But like, I've spent a lot of time browsing
artwork on 99 designs, and on deviant art, and all sorts of random like websites and
like, the stuff I see coming out of this is way better than what the vast majority of
humans, including a lot of like artists who make a living doing art can create an easier
and convenient. And I think it's like, at the end of the day, it's not true that computers
can't be creative, like computers are machines that are almost meant to be creative. And
these things can literally just ingest the entirety of human creation, and a few hours
or days or weeks, and then spit out basically something that is like a result of discerning
all these patterns and what humans can create, and then extending it and going even further.
And I think that like, at the end of the day, like it doesn't really matter if like this,
the lights are on on the computer, if it's continents are sentient or not, right? It's
just a dumb computer, it's none of those things. But it's still spitting out things that I
as like a business owner or as a consumer, look at I'm like, yeah, that's quality art.
Like that looks amazing. Like that's better than anything I could ever hope to paint after
years of study. And so I don't know, I think we're living in like, we're living in the
future. So we've got so. So we've got GPT three that's doing text. So now we've got
digital art. What's next? Well, like, okay, music is a big one. Did you ever see that
movie her with Joaquin Phoenix? That's true. I love that movie. It's got like his AI phone
in his pocket, and like the cameras kind of pointing out from his pocket. So his phone
is essentially following him around in life sees everything he sees, he's got a little
earpiece in and she can like talk to him and like look anything up in an instant give him
all this advice. At one point, there's like a scene he's on the beach and he's feeling
kind of sad. And I think he's just like, oh, play me something melancholy. And she just
like creates this classical composition from scratch to perfectly match his mood. And like
that's another example of computers will never be creative. And it's like, well, what if
what computer can write music based on what it knows you like, from scratch, you know,
like that's coming video, I think it's another big one. Obviously, we're just not getting
to the point where it can do images. But like, it seems like close to perfect. What happens
when computers can make videos, and AI can just basically say, Oh, I want like a video
of me being a superhero and saving somebody, or me using my product and having a great
time. And the computer can basically ingest a bunch of product videos or superhero videos
and create something and interpose your face onto it, right? That's coming at some point
in the future, 3d environments. But right now, if you want to build a video game, you
got to hire hundreds or 1000s of artists who painstakingly craft like every single detail
of the 3d world, what happens when you can tell an AI what you want the world to look
like, and you can describe it, and you come back and it's generated a world, right? And
you can just tell how to tweak it. And GitHub co pilot is an AI trained on I think the entire
entirety of GitHub code bases, which is like millions and millions of code bases. And so
when I'm writing code for any hackers, sometimes it'll just like spit out like, Oh, is this
what you're trying to write? And then we'll just have the next like 20 lines of code written,
just because millions of other people have written somewhat similar code. So it reads
their code reads mine, and it guesses all the variable names exactly what I'm trying
to do. And it's ridiculously scarily accurate. So what is what does this mean? I feel like
you know me, I know you. Neither of us are sort of alarmists, like neither of us feel
like you know, oh, you know, humans are going to get replaced, like both of us, we tend
to find ways to be to just, I don't know, take advantage. And like, it's not it sounds
like a utopian world to me. But what do you think that this means for other people or
for the economy?
I think it depends. It's like that book, crossing the chasm. Anytime there's a new technology
or product that you release, the first people to use it are the innovators. These are the
people who don't care how buggy it is, how shitty it is, how early it is, they can kind
of see ways that they can make use of it to profit or to make their lives better. And
then after that, you got the early adopters, these are the people who are like, still pretty
early, pretty visionary, they come after the innovators, and then eventually get like the
late adopters, etc. Like the people who are like, the last to use anything new, and they
can't understand why it would ever be useful until, you know, everyone else on earth is
used again, then you know, they finally buy a cell phone in like 2012. Right. And I think
that the reality for like this kind of stuff is, if you are a founder, and you are like
looking for a new idea or something, you can probably find ways to use a lot of this stuff,
especially like this new image generating stuff, it's so easy to use, I think the text
is a little bit harder. I think on a societal level, most people won't be directly using
this stuff, like they're not going to try to figure out how to use AI, but they'll be
enjoying the results of the creations that come from it. Right. And so if you could figure
out how to use this stuff to be a tool that augments your workflow, if you're a writer,
and you can generate lots of different texts to like help inspire you or help get your,
you know, your juices flowing, or it'll help you, you know, get a little bit further ahead
on that essay. I think that's awesome. Like this is to me, it's just like a tool that's
going to amplify human abilities. It'll probably put some people out of jobs. But that's been
happening for hundreds of years. In the 1800s, almost everybody used to be a farmer. And
then we got like farm equipment, and suddenly, you know, you don't need so many farmers.
That doesn't mean that everybody is just like unemployed. That means people sort of rise
to higher callings. And so what I like to hope the future consists of is that we have
fewer and fewer people doing sort of old jobs, and we have more and more people doing jobs
that are like a little bit more creative, a little bit more fulfilling, a little bit
more personal, a little bit harder for computers and machines to basically replicate, which
we're already seeing, you know, and there's a dystopian version of that, like the sort
of gig economy, Uber driver, everybody is just like this, you know, person you summon
from an app works for pennies. But it's kind of a utopian version of that too, which is
people instead of being gig economy workers are startup founders, and they're self employed.
And they essentially, what Legion calls the passion economy, they're following their passion,
they're getting paid to do it. Whereas like, you know, the year 1850, no one's gonna pay
you to like make tiktok videos, because number one, tiktok doesn't exist. But number two,
like get to work, like we need to eat, right. And so I think we, we might be heading toward
a society where people can do more fulfilling things, because sort of the grunt work of
just doing basic illustrations, and writing basic essays, and, you know, writing code
or whatever gets automated out. And we need more creative tasks.
There's a famous quote from I think it was john Maynard Keynes, back in like the 1920s
or so. And he he made two predictions based on the trends in technology, the way like
the speed of innovation at that time. And only one of his two predictions came true.
One of them was like, by 1990, there's going to be so much wealth, there's going to be
so much, like, you know, technological advancement, that we're not going to have to work that
much anymore, as much as we have to work here in like the 1920s. And indeed, we're all gonna,
you know, we're all going to have, you know, these like three, three day work weeks. And
you know, everyone's going to be doing exactly what you just said, spending a lot of time
doing creative things and whatnot. And his technological predictions came true is like
wealth creation predictions came true, we really do live in this in this world. I mean,
I find Dolly is really cool. But I mean, just the crazy pace of innovation in the last 20,
30, 40 years has been staggering. But the second piece didn't come true. Like we're
still working as much as ever. We're working perhaps more than ever.
The quote that comes to mind for me is from William Gibson, it's the future is here, but
it's not evenly distributed. Right? Like if you're in San Francisco, like I just saw a
story, cop pulls over driverless car. So cruise technology has their driverless cars. And
they've literally are now I think legal in San Francisco. And they just drive around
like driverless Ubers, and they can pick you up, you can summon them, they'll pick you
up, take you where you want to go. There's not a driver in the car. And one of these
cars had the headlights out. And a cop pulled it over because the headlights were out. And
then people started filming this, like, Oh, sure, what's gonna happen? Like, it's on a
busy San Francisco street, and they see this cop walking into the car. And like, presumably
the officer didn't even know there wasn't a driver, he walks up, and there's no one
in the driver's seat. So he's like trying to figure out like, how do I open the door
and turn the lights on, etc. And the car just like pulls away. Because AI is like, I'm not
parked in a safe spot, like I'm gonna pull over here. And who do you give a ticket to?
Right? Like, that's a very futuristic problem that like, I couldn't even have imagined 20
years ago. But if you live in San Francisco right now, you're like recording that on your
phone. And that's happening, huh? I'm reading this book right now, by this author, who's
a game designer. And it's about literally how she thinks that we should try to gamify
the world. And one of the things that she points out is that we like to have rules.
And we kind of like simple game worlds, where everything is kind of simple, it's clear what
the the action steps are that you should take. There are two sides to a lot of new things
and a lot of things changing. On the one hand, on the one side, people who are, take a lot
of initiative who kind of have an entrepreneurial mindset or an early adopter mindset, have
a field day with it. Because I think in a lot of ways, you have to be the kind of person
who says, Oh, hey, like there's new stuff in the environment. I can't wait to play with
it and combine it and recombine it and figure out how I can like, you know, like make it
work for me. Yeah, agreed. I like doing this with you. Let's do more of these. I'm definitely
down to do it. Just so that I can keep you accountable and make sure that the podcast
is staying up to the right kind of standard. Yeah. All right. Later, dude. All right, later.