Erik
Hello, welcome to Picture Me Coating with Eric Aker and Mike Mull. Hi, Mike. Welcome to your show.

Mike
Thank you. I appreciate it.

Erik
Mike, I have been listening to a record this week that I'm sure you will not be surprised to hear me mention to you here because I keep telling you about it. The band is called Friends in Real Life and the record is of the same name. When I first heard it I was like oh this is kind of fun. It's not brilliant. It's not like heart-rending. It's not gonna I'm not like gonna fall in love with it but the more I listen to it the more it's like endeared itself to me. A lot of what he says on the record makes me laugh. The guy who wrote it the guy who is this Friends in Real Life band used to record under the name Pat the Bunny. It is his background as like a gutter punk, anarchist punk type of guy. But then he fell pretty hard into drugs and alcohol and had to get out of that. And then now he's in this, he's got this group called Friends in Real Life and he's kind of like emerged on the other side of, you know, more like grown up adult, not gutter punk anarchist, but just making music. I guess he started just carrying a acoustic guitar around because it was easier to play these protest songs he had without having a huge backing band. You just have your guitar and play for people. Anyway, there's a song in there called Surf Rock Anthem Lo-Fi Steady Beats Remix, which itself is pretty funny. And he says, if I die young, say it was the government. Even though it wasn't, give the kids something to believe in. That's a lot of stuff that makes sense.

Mike
That's a great line. Great line.

Erik
But he actually worked as a software engineer. And he talks about it on this record, too. There's a song called Retirement Plans. he's talked about his parents and his working life he says he loved working in software and then he says it was ruby on rails in the back end i know it ain't hip and i hear that i wonder why he said

Mike
that i wonder if he would like to be a podcast guest i don't think he does a lot of interviews

Erik
he's sort of like a reluctant musician all these years i think maybe he's saying from the perspective of the gutter punks it's not really hip to be going to work every day and typing ruby on rails I guess he got injured, though. He says in that same song, he says, we wrote code for social workers, and he liked it. But then he says, not going to fix a broken system, but I don't know what is. Some days he just started waking up in pain, and he wasn't good for all that typing. And he says, sometimes the things you love, well, they just don't love you.

Mike
I listened to this a couple of times, and I kind of liked it. I hear hints of other things, you know, like maybe bits of the violent fems or some touches of they might be giants or something like that. You compare it to something else that I can't remember now.

Erik
Against me.

Mike
Against me, yeah. Some of the sounds sound a bit like demo tapes, but

Erik
it's a fun listen. I think I'm a sucker for it because I relate to a lot of the stuff that he's singing about.

Mike
I think it probably helps to be a surfer.

Erik
There's this kinship with the gutter punks and the hardcore scene. I remember when I first started working as a software engineer, I would feel like such a weirdo going into offices. Like, wow, I really don't belong in an office. These people, if they knew where I came from, they would be like, you just don't belong here. You should get out of here. Imposter feeling. But sometimes I would go to lunch with people I worked with and we would just start telling stories. I remember one day I was telling the story, it was probably a story about how there was this big brawl at a hardcore show. And the response that I got from someone was like, wow, you've had a weird life. And it was like, oh shit, I shared too much. They found out I am kind of weird.

Mike
Yeah, you have a lot of sort of like Quentin Tarantino side quest stories.

Erik
I don't see what it was like.

Mike
I went to Mexico with my buddy in a car that he built himself.

Erik
I don't know. I don't feel that way. When I first started working in the business, quote unquote, business world, though, I did feel like an oddball. I did feel like, wow, I don't belong here. I think that's what he means when he's saying, I worked at Ruby on Rails. I know he's thinking about his friends he used to hang out with and, you know, punk squats and hanging out on sidewalks, you know, trying to beg for spare change.

Mike
I was flipping through music sites and I came across this headline that was something like Norwegian Doom Band Cryptograph has new album Kryptonimacan

Erik
Wow, that sounds right up our alley,

Mike
doesn't it? I was like, I haven't heard it yet but I think there's about a 99% chance I'm going to love this album

Erik
Wow, that's a great chance

Mike
Just based on the description It was a little surprising. So I was thinking like more Doomy. It's a modern Norwegian Doom band. So I was really expecting something more like Paul Bear or, you know, even Sleep or something like that. It turns out the music here is kind of more along the lines of late 60s, early 70s, hard rock, psychedelia type stuff.

Mike
But it is good. It's interesting. It's got this sort of like nautical theme.

Mike
Album cover is really cool, too. It's got this sort of octopus squid thing on it. And there's this sort of through line of like a nautical journey. So there's songs called Beyond the Horizon, Lost at Sea, and The Gales. Oh, okay. Based on the description and the cover of the album, I was kind of expecting something to supplement The Sword, one of our favorite Friday afternoon Get in the Zone albums. And it's not quite that, but it's a fun listen. and I really like this sort of retro hard rock sound that it's got.

Erik
Yeah, that reminds me, your description reminds me of Graveyard. I felt almost a little guilty listening to Graveyard because it felt almost like a pastiche, but I loved that one record that they put out. Yeah, there's a little bit of that

Mike
here too. There's cases where it sounds like they're trying to sound like Hendrix or trying to sound like Cream or trying to sound like Sabbath or whatever, but I don't know how it works somehow.

Erik
Well, it's like Black Sabbath went into the mines of Moria. I'm going to mix the metaphors here, right? And they started digging down for the mithril, and then they stopped. And you can't help but look in there in that cave and go, well, wait a minute. There's such a rich vein of material yet to mine. Somebody's got to go in there and see where it goes. So when these fans do that, I'm like, yes, they're following that. lucrative strain. Yeah, if

Mike
nothing else, it inspires you to go back and listen to the sources. It's a lot of fun. With a name like Cryptonomicon, which a lot of our listeners probably recognize as the title of a Neil Stevenson book, it just felt like something I had to mention on the show.

Erik
I've got to check that one out. Honestly, it sounds right up my alley. Mike, this week, I have had this problem where I've been walking around for about two weeks I keep seeing these news stories and I keep thinking, oh, I should tell Mike about this. I should tell Mike about this. And so after I collected at Epivum, I was like, okay, do you want to do a news episode? I know we do this every five, six months. I don't think we're the greatest talking heads for here. Let us interpret the computing technology news for you. But there's weird stuff that happens. And I feel like somebody has to raise it. Somebody has to talk about it. And who better than us, Mike?

Mike
I certainly can't think of anyone better.

Erik
the first article that i saw that i came across mike and i was like i gotta tell mike about this was was about the the humanoid robots running the half marathon in beijing i'm sure you heard about this story did you hear about this story yeah

Mike
it was pretty widely disseminated i only recently read the articles and looked at the videos when i first saw the articles i was like you know so what big deal. You know, I'm an old man. Almost everybody beats me in the half marathons now anyway. So it doesn't bother me if robots beat me. But I've

Erik
thought of you because you are a runner. You've run a lot of marathons and half marathons and you are into robots taking our jobs. Someday these robots are going to be running marathons for us and all the marathon runners are going to be out work. I was somewhat encouraged

Mike
when I started looking

Erik
at the videos.

Mike
The best time for the robot, for a robot, was about 240. Yeah. I don't want to insult anyone out there because running a half marathon is an admirable thing to do under any circumstances, but 240 is pretty slow for a half. 240 sucks, is what you're saying. Yeah, I'm pretty encouraged by the fact that even at my advanced age, I could kick a robot's ass and have marathons.

Erik
So the rules, if you wanted to enter one of these in the future, Mike, they have a website. It looks like a website from 20 years ago. It's called the Humanoid Robot Marathon. The rules to enter these robots into the race are it has to resemble a human, which is a pretty fuzzy rule. It has to be able to walk or run, so you can't have any wheels. And there were 21 robots that competed. Only six of the 21 actually finished the race. As you mentioned, the top time for the robots was two hours, 40 minutes. And they gave that robot a little gold medal that was robot shaped.

Mike
Yeah, I don't know what to think about this story. It's kind of amusing to watch them. And I guess it's interesting to see, you know, people talk about how, wow, this is amazing because, you know, a decade ago, we couldn't figure out how to make robots walk like humans. And

Erik
now

Mike
they are arguably running. So it's an impressive advance. But it wasn't a situation where robots are going to be beating humans.

Erik
So the reason I wanted to bring this to your attention is I read like half a dozen articles about it. And the theme that kept coming up in article after article was, look how sad and funny these robots look when they're trying to run a half marathon. So there's this automatic impulse to anthropomorphize them, to see them as like sad imitations of people. And then there was also the humans don't have anything to fear anytime soon. So it's not like Casparov and Deep Blue. It's not like robots are going to come and start beating humans in marathon races. I was surprised by how common these two themes were. What you said was, oh, the technical feat of going from making a robot walk to making a run to making it run on half a marathon, that mostly wasn't cut. It was like this is a novelty story in the news. And I think that there's a hint of how we see technological advancements. Like there's the whole story of technological advancement as it gets described is present in this story too. It's like, oh, look at this funny thing that these weirdo researchers are doing. Look how unthreatening and lame it is.

Erik
And then suddenly it becomes, oh, wait a minute, it's beating us in jeopardy in chess and go. And oh, wait a minute, it's writing code for us too. There's this crossover moment. I think the funny thing about this one is the crossover moment doesn't seem threatening. Like if the robots beat the people in the marathon, is that as interesting as robots beating humans in chess?

Mike
I've never been able to find the original source for it, but I remember reading a while back, like decades ago, that somebody was interviewing Isaac Asimov. And, you know, this was way before the days of the original deep neural network stuff. But somebody was interviewing Asimov and said, do you ever think we will have robots that are just like humans? And his response was something along the lines of, I hope not, we already have humans. And I always think about that when they come out with stories like this, because personally, I don't want robots that can run half marathons. It's an interesting technical feat, but I don't care if robots are faster than humans. I don't care if they're slower than humans, whatever. It's just not something I want from a robot. You know, if I, from a practical perspective, you know, a robot that will do my dishes or, you know, grow food for me, those are interesting things. If you can show me a robot that can, like, plant tomatoes for me, then I will probably get enthused about that. But a robot that can beat me in a half marathon is not going to

Erik
excite me or perturb me. Not even if you imagine strapping weapons to it and having it hunt down your enemies?

Mike
Yeah,

Erik
well, you know, if it's a robot

Mike
that can run at 30 miles an hour without stopping for 13 miles and is carrying a weapon, then it will definitely terrify me. But, again, that's not something I particularly want.

Erik
For the Asimov quote, I often think if you had a truly autonomous robot, would it choose to play chess or Jeopardy or run a half marathon? Or would it choose some other activity that we can't foresee?

Mike
Yeah, it's a good question. It seems unlikely that they will be set up to obey Asimov's laws, because it seems like robot soldiers is the inevitable outcome of human-like robots.

Erik
Right. We have to have some practical application, right, Mike? And that's the easiest one for our minds to leap to, for mine anyway. So that was the first story I wanted to tell you about. What did you read about this week? So this is a relatively

Mike
small-scale story, but I thought it was interesting, and I am surprised that it didn't get more tech news coverage. So there's the Cloud Native Computing Foundation, CNCF, which has been around for a while. Basically, they act kind of like Apache for cloud-native projects, the idea being that open-source projects, which may have come out of academia or industry or whatever, are are sponsored by this foundation and then developed as open source projects going forward. You know, it's a nice idea to get important open source projects funded and supported and exposed to the public and so forth. So


Erik
I usually associate CNCF with Kubernetes because they sponsor the KubeCon and all. There's a lot of Kubernetes adjacent projects that are CNCF projects.

Mike
Yeah, Kubernetes and Kubernetes-related things are probably the things most associated with them. And, you know, huge successes and an important part of our ecosystem. So anyway, I came across this article. They had published a thing on their website. So the title of it is Protecting NATS and the Integrity of Open Source, CNCF's Commitment to the Community. I confess I had to look up what NATS is. Me too. It's a little bit hard to describe, but it's kind of this very sophisticated messaging system. You can do like PubSub and streaming, and you can do this sort of like named endpoint thing. So people, I guess, use it for like multi-cloud deployments and edge computing and things like that. It was originally developed by this company called Cinedia. And Cinedia in 2018 basically gave this project to the CNCF. It's been open source for close to seven years now. So the heart of the story is Cinedia now wants it back. Wow, you can do that? Well, what does

Erik
it mean wants

Mike
it back? So what they say in the CNCF article is Cinedia, the original donor of the Nats project, has notified CNCF of its intention to withdraw the NATS project from the foundation and relicense the code under the Business Source License, the USL, a non-open source license that restricts user freedoms and undermines years of open development.

Erik
Wow. So the CNCF needs some kind of like jinx, no take backs type of way of receiving these projects.

Mike
Yeah, and I think they go into detail in this article about how Cinedia had made verbal commitments to transfer various bits of IP to them. And I guess things didn't happen the way they were supposed to. But it's interesting to me because it seems like it presents an interesting precedent. So if Cinedia is successful in withdrawing an open source project from CNCF, it seems like a bad sign for the future of open source because, you know, it seems like any company could get to a point where a project is popular and, you know, has a significant user base and then say, whoopsies, we meant to make money off that after all.

Erik
Yeah, it's kind of like, hey, can you foster care my kid through the most expensive years of his life, high school and college? When he gets done, I'll take him back.

Mike
Yeah, right, exactly. Once he's in the workforce, I'd like to have access to him again. Once he's earning. So anyway, I guess we'll see what happens there. It's likely to be wrapped up in legal arguments for a while.

Erik
I hadn't heard about this. It again raises these oddities for me of what open source is like. These companies donate software that they've invested a lot of time and money into, and then outside contributors come and contribute, and then the companies are trying to figure out how to make money off of that software, but the whole industry that we work on is built on this massive foundation of freely given away work. There's a sort of paradoxical energy somehow at the root of what we do.

Mike
It's not a fun story, but I thought it was significant. And since it didn't seem to get a lot of traction in the tech press, I thought I would bring it up.

Erik
If someone ever donates an open source software project to you and I, Mike, as part of the Picture and Recoding Show, we'll have to say no take backs.

Mike
Yeah, we should look into the no take backs licensing.

Erik
I read this weird article this week that I don't know how relevant it is for what we normally talk about, but I just kept thinking about it. I wanted to share with you. It was from the BBC, actually. It was called Health Bars and Power-Ups, the, quote, freaky and unpleasant world when video games leak into the physical realm. So this article was about how people mentally transfer elements of video games into the world around them and how virtual experience goes over into reality. In the piece, there's a researcher. Her name is Angelica Ortiz de Gortari. She coins this term, Game Transference Phenomenon, or GTP. Game Transference Phenomenon. And here's from the article. One study participant reported seeing health indicator bars like those in the role-playing game World of Warcraft floating above their companion's head. Another spoke of lapses in concentration after not being able to stop seeing images from a game. others said colors in the real world seemed transformed and began to mimic the colors of the game world they had recently played in now the researcher says the more realistic a video game world is the more likely players are to confuse the game world and the real world so this is a weird sounding phenomenon but it struck me because i've been playing a lot of tetris lately and sometimes when i talk to people i kind of zone out start imagining these tetris blocks they're called Tetrimino's Falling. I'm organizing them in my head. Gortari says, if a video game is realistic, you're more likely to confuse a game in the real world. So my question for you is, would you say that Tetris is, quote, realistic?

Mike
I mean, I find myself trying to drop bricks into optimal places all the time in the real world.

Erik
I don't know if you've answered. I don't know. I don't find Tetris to be really realistic, But I feel bad when I'm talking to somebody and the Tetris pops in my head. And I'm not getting credit for clearing all these lines mentally that I'm clearing.

Mike
Is there like a 3D version of Tetris? Is that just Minecraft?

Erik
I don't know, actually. Minecraft, I think you're more breaking. I guess you're building in Minecraft. So the two things that confuse me in this article. They say it's uncommon, but it may also be common and it might be dangerous. I don't understand how it's dangerous. Is it a form of hallucination? The rare versus the common thing appears in this paragraph. So she says that the researcher talked to gamers in different studies. The largest study she published was in 2024, and it involved the participation of 623 gamers, male and female. The results suggest that between 82% and 96% of those gamers have experienced some form of GTP, game transference phenomenon. Are you surprised to hear that this might exist?

Mike
I'm a little surprised by it. There's also this aspect to it which makes me think, I hope the augmented reality people don't hear about this. Because then there's this likelihood that they will start trying to induce this phenomenon rather than trying to mitigate

Erik
it. I guess I was surprised that this exists. The article talks about how if you play games for long stretches, two, three, four hours or more, you're more likely to experience this phenomenon. When I read the article, I was like, well, that happens to other people. I was confused about what the danger is. They describe it as an involuntary change of perception. This might be it. Somewhere in the article they write, studies from Guattari suggest that GTP induces distress and dysfunction for around half the gamers who have experienced it, with confusion, hypervigilance, and irrationality among the symptoms.

Mike
That might explain a lot of things.

Erik
It's kind of a funny sounding story, but I also thought, whoa, that happened to me and I felt bad about it. And this is a thing that happens to other people. I was surprised too that I would have thought this is much more common with like virtual reality, as you pointed out, augmented reality. That you could play a video game and then just sort of transfer over to the world. My wife always talks about this game Katamari where you roll this giant ball around and you blob up like messy objects in the world and you end up like cleaning the world. And she will sometimes be in a place and there'll be a lot of visual information, I don't know, grocery store, whatever. And she'll say, I wish I could roll all this stuff up. That might be a dream transference phenomenon. GTP, I'll tell her. Yeah,

Mike
I don't know. If you're in the produce section at the grocery store and you start seeing Pokemon, that seems like it's just a straight mental health issue, right? I mean, that doesn't seem like it's

Erik
innocuous.

Mike
But if you know

Erik
that it's game trends, if you know, not necessarily the name for it, if you know, oh, I play that game a lot, so I'm seeing those things everywhere. If you know that, it's not the same as straight mental health. It's the transference of this mental imagery from outside, from the game to the outside world.

Mike
Right. But if you had taken, you know, psilocybin and then, you know, a week later you start hallucinating things,
you would probably consider that to be a medical problem that you need to get fixed, right? You're saying I need to get fixed for this Tetris

Erik
thing.

Mike
Maybe play with Tetris. Tetris? Yeah,

Erik
I don't know. I've been trying to clear out work out of my brain lately, and for some reason that's one way to do it. Just sort of like take a hammer and smash those work thoughts out of the brain. I think the answer is forest bathing oh okay you go find pools of water in the forest and you bathe in those that's the answer huh

Mike
not exactly but you need to get you need to get out into a natural setting and completely unplug no phone no headphones

Erik
just listen to the stuff

Mike
around you

Erik
can I do that for years how do I make money if I do that for years um you probably would object to

Mike
hunting, so you'd have to become a forager. Okay. What have you got? What else did you read about? So, again, this is a relatively small-scale story, but it definitely made me laugh out loud as kids say. So, there's this site called Retraction Watch that literally keeps track of retracted scientific papers and stuff like that.

Erik
Oh, man. I hope to never appear on that site, but the chances of me appearing on there are pretty low. Well, you might think that,

Mike
but let me read this paragraph from the site. In March 2025, sleuth Guillaume Cabanac, creator of the problematic paper screener, pointed out in a Pubpeer comment that paper included several tortured phrases. So he's talking about this AI paper that was written by a Google engineer. These phrases indicate AI use and occur when large language models try to find synonyms for common phrases. In the paper, linear regression became straight relapse. And
error rate became blunder rate. That is pretty good. My personal favorite here, too, is the phrase mean squared error is used pretty commonly in machine learning papers, usually the abbreviated MSC. But because the AI was trying to be original, it in the paper called it mean squared blunder. And if I ever do write an AI paper in the future, I'm definitely going to call it mean squared blunder going forward.

Erik
That's a life imitates art. But then how do people know that you are a real human writing that and not the AI writing it for you? I think you have to like footnote it or something.

This is a hilarious reference.

Mike
Yeah. I'm trying to be funny. I am not an AI.

Erik
Oh, I like straight readouts, but it's so hard for me to connect that to linear regression. It just sounds like a hilarious pun.

Mike
Yeah. Anyway, this paper was written by an engineer at Google named Anurag Awasthi. And, you know, he tried to make some comments around, you know, he was doing an experiment and kind of got away from them or whatever. But I thought it was interesting for two reasons. One, it kind of points up the perils of using AI in your writing, especially if you don't pay close enough attention because it will do things you don't expect. And the other one was just the, I just thought it was hilarious. The particular changes that it introduced are sort of comical. Mean square blunder

Erik
is pretty good. So the story that I read that I most wanted to tell you about was, I read it in the print copy of The New Yorker. It was a long-form essay comparing the Luddites, the period of the Luddites in 18th century and early 19th century England to the current AI revolution. And the title of the article was Luddite Lessons, How to Survive the AI Revolution. I think on the website version of the article, on the New York website, it's just called How to Survive the AI Revolution. So this article, they start by giving context for Luddite battles in England in the 19th century. I didn't really know much about this. Do you know anything about this? Do you recall the historical context around the Luddites? Putting it on the spot

Mike
here. Just vaguely that it was sort of a reaction to people automating jobs, which had previously been done by humans at the beginning of the Industrial Revolution, you know, and the obvious parallels. You know, this term has come up a lot throughout my life.

Erik
Me too, yeah. I had no context for it virtually. And anybody who's opposed

Mike
to the latest technological breakthroughs is inevitably described as a Luddite. So, you know, I'd always heard that it was related to this guy Ned Ludd, although I think this article particular sort of questions that origin. Yeah,

Erik
they say that people aren't really sure where the term Luddite comes from. So these were textile workers, and they protested and rioted and destroyed these automated textile machines. Like there was this one that was steam-powered, and it wove cotton thread into fabrics. This was all like hand-done before the end of the 18th century. Parliament passed laws against these people, like breaking these, they called them frame breakers, who break these machines, and they hanged nine people for breaking these machines in the riots. The article says that they were, the people who engaged in this work were like a priesthood, that they would pass this knowledge down, cultivated, specialized knowledge. There's a line in the article that the Luddites grasped an essential truth. The factory system threatened their artisanal economy and livelihoods. And so this is my question for you. Would you say software engineering is an artisanal economy or livelihood?

Mike
There's certainly some parallels. I think it has become less that way in the last decade or two. But, you know, there's parallels in the sense that my assumption is when these people saw these machines taking their jobs, their initial reaction was, you know, how could a machine conceivably do this delicate and sophisticated thing that I have learned

Mike
and had passed down to me through generations? And then the other parallel that they talk about in the article that basically you can think of what's happening now as a way for people who own the companies to sort of take back control of the labor because there's been this long, especially in the tech industry, there's been this long period of where the labor of people producing the software and producing the technology have been sort of in control of the economics of the industry. labor pool. And the ability of those people to sort of take that back across the white-collar labor pool is very similar to what was happening in the early stages of the Industrial Revolution.

Erik
The comparison is fascinating. In the article, they talk about how this affected a variety of textile workers, but it was most apparent in cotton manufacturing. When they would turn cotton into finished fabrics. It had two stages. You first have to take the fibers from the cotton and then spin it into yarn. And then you have to take that yarn and turn it into, weave it into fabric. And so they had actually mechanized the first part, the spinning part. So they had machines to do that. But the second part, weaving it was too complex to figure out how to automate it first. And the article they write, this technological imbalance actually increased demand for hand-to-room weavers, whose numbers grew in Britain from 37,000 to 208,000 between 1780 and 1812. These weavers, many of whom worked from their homes, enjoyed rising income until about 1800. So you know what that
reminded me of?

Erik
I was thinking, wow, there's like kind of two components here. And when you automate one component, it suddenly blows up the demand for people working in the other component. It was almost like in software how once it got a lot easier for us to deploy and scale our software products, you suddenly needed a lot more people to produce the stuff for all kinds of organizations who want to easily deploy it and scale it. What do you think about it? Is that kind of loony, that comparison? I

Mike
don't think it is. So there's a couple of things here. I think people in the software industry have been largely unsympathetic to the idea of people losing their manufacturing jobs. And I think one reason is because, you know, in the back of our minds, we've had this notion that technological progress always brings around more labor. You know, there's this sort of Jevons paradox thing where technological advances have the odd effect of actually producing more or necessitating more labor. And so I think as technology people, we've always been kind of unsympathetic to the idea of people being replaced by robots or people being replaced by efficiency improvements or mechanization or whatever. Now there's this sort of thing where we're like, hey, wait a second. Yeah. We didn't mean us. Yeah, we didn't mean us.

Mike
It's interesting because you can sort of intellectualize this idea that technological advances always eventually mean more labor. But when you're caught up in the reality of it and realizing that nobody's quite clear what that new labor is going to be and how quickly it's going to come about and if it's going to come about and what the transition is going to be like, it is very worrisome.

Erik
They talk about that in this article. There's this sort of nonchalant perspective from economists. The article mostly quotes the work of economists and interviews, the writer in the article mostly quotes economists, interviews economists, and talks about their work. The history lesson about the Luddites is all about this preamble of talking about AI. And there's this nonchalance that certain economists have, which is like, yeah, every time you lose jobs, you gain them somewhere else. But if you're the person on the losing side, it doesn't maybe feel like so great that there's this overall economic rebalancing that's happening.

Mike
Yeah. Also, one of the clear parallels here is I assume these people, like the weavers, assumed they were, you know, skilled labor and they were doing something that was, you know, a craft of sorts. And probably people were annoyed by losing their coal mining jobs too, but it's probably a different thing because maybe you don't think of coal mining as being something you want your children to do for

Erik
your grandchildren to do. Yeah, that's a

Mike
good point.

Erik
It's an unsettling comparison, I think. I hear this comparison and I think, ooh, that sounds apt, but it also unsettles me. The writer of the article says that this Luddites era, comparing it to AI is apt. You see this comparison given even by people who are techno-optimists, who are saying, this is great, but it reminds me of the Luddites. So the piece, as I said, is mostly about economics, and most of the work talks about the work of economists. There's also other articles, books, and podcasts about the Luddite period and comparing it to the current period. In this article, they interview a number of economists, including a couple from MIT. So here's a line from the article. When AI automates human skills outright, MIT economist Eric Brydjolfsson warns, quote, machines become better substitutes for human labor, while workers lose economic and political bargaining power and become increasingly dependent on those who control the technology. In this environment, tech giants, which own and develop AI, accumulate vast wealth and power, while most workers are left without leverage or a path to improving their conditions. Bryn Yelpsen termed this dystopian outcome, quote, the touring trap.

Mike
Yeah.

Erik
This reminded me of something you had said about how it would make you sad to imagine that maybe we're using LMs in the future, but they're like you go and you rent them. So if you want to produce software in the future, you have to go and pay for time, beg for time from some large company that helps people produce it. I don't like this phrase a lot. I don't either. That's why I was like, I got to give Mike this quote. What do we think about the

Mike
name Touring Trap? Yeah, I don't love it.

Mike
An alternative does not spring to mind. Why don't you like it? Well, first of all, I think it's unfair to blame Turing for the predicament we're in. Yeah. Even if he is somewhat associated with AI. Secondly, it just, it doesn't mean anything to me.

Erik
There's nothing specific in Alan Turing's work that says, oh, you know, this is how you can enclose unwilling
participants in this complexity trap where they need your computation to survive.

Mike
Yeah, you know, if somebody were to ask me, have you heard about the touring trap, I probably would not immediately spring to this idea, you know.

Erik
What about the critique, though, that these large companies accumulate wealth and power and they own and develop these systems and everybody else has no way, no leverage, no way to either resist or change their economic conditions?

Mike
I do worry about that. I mean, I think it's part of a broader social phenomenon. This
is a bit of a diversion, but I was listening to Gavin Newsom's podcast this last week, and yes, I confess, I listen to Gavin Newsom's podcast. Feel free to ridicule me. He had Scott Galloway on his podcast, who I'd never really heard of, but is apparently a fairly well-known cultural critic. Anyway, Galloway was talking about how he got into the Berkeley Haas School, the business school of Berkeley, with a 2.27 GPA or something like that,
back when he applied in the 1980s. And when he got out of business school with MBA, he was able to buy a house in the Bay Area for like $280,000. And in the meantime, the salaries that you can get as a Haas graduate have gone up considerably because it's a prestigious place. But they've gone up like twice, while the housing prices have gone up 10 times. The idea being that if you're a young person now entering the labor market, it's probably not going to be realistic for you to buy a house in a lot of areas. So homeownership is kind of beyond people. The idea of getting married and having children is kind of daunting and economically difficult now. So, you know, there's been all this stuff over the last decade about how people are buying houses so that they can rent them out to people. And it just feels like there's this more and more focus on people having to rent things. Like, I grew up in the sort of personal computer era. And so the idea of being able to control all of your technology and do everything locally is very appealing to me. But it feels like there's this risk of getting into this situation where you sort of have to be dependent on these cloud providers or the tools that they're building. Because the infrastructure that you need to improve them and do inference with them is just beyond the capability of people to purchase and run locally.

Erik
I think that's the threat, that we come to rely on these tools. We can't perform without them, but they're owned and operated by massively powerful companies. And so you have to beg for the pennies to, or the units of time, to be able to rent that tool to perform whatever your work is in the future. The journalist who wrote this piece talked to another MIT economist, David Auteur, who brought up this 2023 study. The study showed that, quote, certain highly trained radiologists when using AI tools produced diagnoses that were less accurate, in part because they gave too much weight to inaccurate AI results. This is a quote in here. The tool itself is very good, yet doctors perform worse with it. This actually happened to me this week. I was shipping this throwaway script, and it was going to affect production behavior. I just needed this script to run a couple times. I needed to run this script a couple times and prod and collect some data and stuff. So I mostly let Copilot wrote it, and I didn't even really read. And it was even PR. There was a reviewer for my PR. I let Copilot take the wheel, and it called all these functions with the wrong arguments. There were functions I had written. I just didn't even really read closely. And so I read this quote in this article. I was like, holy shit, I did that. I did a bad job this week because I didn't pay attention to what Copilot was writing. Why did I do that? I don't know. I think I would have taken it a little more seriously if I knew it was going to affect production behavior. or maybe cause an outage in this kind of what I consider to be a throwaway scenario. I was like, oh, whatever, it's probably fine. And I shipped it, and I merged it, and somebody reviewed it, and it was wrong, and it did not work. And I thought, wow, here I am doing worse work. The tool is very good, but I may perform worse

Mike
with it. Yeah, but, I mean, it

Erik
was wrong,

Mike
but think of the sure volume of what you produced. I mean, could you have produced that error as quickly in the past

Erik
without these tools? That's quite a nice point. I probably could not have produced that error as quickly. The article, I was a little disappointed in this article. It was interesting. I liked all the historical context about the Luddites. I didn't know the stuff about the Luddites. Of course, I've heard the term thousands of times. I liked the comparison to our current era. It seems like a fitting comparison to me. I like trying to imagine these textile workers, their artisanal craft, their priesthood of passing down this knowledge to their children. It's a safe job at the time. It's a job that's useful and productive. And then it was completely mechanized away. So I thought this article is interesting. Maybe we can learn from this moment in time, the Luddites. And the article just kind of ends like, yeah, so this is a little scary. And they close on a quote from the economist referenced above, David Auteur from MIT. He actually says, in his opinion, quote, we could have handled this challenge better in the 1970s than we are now. Ironically, I have less faith in our ability to manage it today than I would have had when we were a lower tech society. This is probably a bad moment for AI to appear. That's the final line in the article. And the reason they end with that is it's like, well, if you want to avoid massive disruption in your society, You probably need to pass a bunch of regulation, and you probably need to sort of guide the cultivation and development of this tool so that it's used in this way where you can still have knowledge workers making decisions. We're not going to do any of that, so this is probably a scary moment in time for us. What do you think about that? The 70s would have been a better time for this? That might be a

Mike
little bit of rose-colored glasses. I think that the 70s were a difficult time for different reasons. I do have this thought fairly often that I don't think people understand how weird the world is right now. And I don't think they understand how much weirder it's going to get really beyond our capacity to adapt to it. So I'm sympathetic to the idea from that perspective, you know, that the replacement of lots of white collar work with AI agents and the emergence of other countries as technology powerhouses that maybe we in the United States are now behind. It's just going to be a weird time, and I wish my kids were not going to be the people living through these interesting times.

Erik
Yeah, interesting times. It seems like extreme uncertainty ahead. I guess we've talked a lot about the AI revolution. We've talked a lot about LLMs. And as a practitioner, it feels like, yeah, okay, this is something that potentially threatens my livelihood. Potentially. It may be coming. But there's this wistful, sad part, too, because it's not just a livelihood. This is actually work that I enjoy doing. We talk about this, of course, many times. So I have this job that I also happen to like working in. And here's the thing that threatens both the livelihood and the enjoyment, that kind of satisfaction that I get from the work itself. The final piece I have for you, Mike, was about your old company, Yahoo. I brought this one to you. I read this and I was like, oh, I got to send this to Mike. Yahoo wants to buy Google Chrome. This was covered in The Verge. I think it was probably covered all over the place because people were kind of laughing. I think Yahoo is going to own Chrome. That'll be the day. That's hilarious. So what happened was, and I'm sure you are familiar with the story, but just to recap, Google was sued by the U.S. Department of Justice in 2023 for monopolistic practices. And this lawsuit alleged that they have this dominant market position to set ad tech costs and parameters around ad tech. They're kind of the 800-pound gorilla, right? And the reason that they have that is that they control so many of these places that have a synergistic relationship where ads may appear. So, for example, Google Chrome, because most of the time when people want to search a thing, they just type it into the Chrome browser bar. But then they also have Google Maps. They also have YouTube. They have other apps. Google Maps, for instance, when there started being ads in Google Maps, like really annoyed me. Because, you know, I have a disease that prevents me from ever even considering going to eat at a Wendy's. And yet Google always insists, oh, you're going to this place. I'm going to give you directions. It's right by the Wendy's. Didn't you want to know there's a Wendy's over there? You could also stop by the Wendy's. And I'm like, Google, stop telling me about the Wendy's. I don't care. So it's just like subtle dystopian ads leaking into something that for me has always felt really purely informational. I just want to look at the map and know where the things are. I could care less where the Wendy's are, Google.

Mike
I kind of want french fries now, though.

Erik
So they actually, the lawsuit was decided, and they determined, there were two actually decisions in these suits, that the company, that Google has been using its, abusing its power to stifle competition. And the technology that they use for advertisers to get ads in front of consumers that they have unfair monopolistic practices around those. So now that it's been decided, they started this three-week remedies trial, and the Department of Justice has all these suggestions for the remedies trial for what to do to remediate this monopoly that they claim Google has, that the lawsuit determined Google has. One of the remedies is they should sell Chrome. And so then they had, at the trial, they had all these people at the remedies trial. They had people from companies saying, yes, yes, we would buy Chrome. And that included Yahoo.

Mike
I have

Erik
somewhat

Mike
mixed feelings about this, I guess. I use Chrome pretty regularly. I know that a lot of people in my circle, they prefer Firefox for their personal browsing. And I know even though people who have switched to DuckDuckGo, both for the browser and for their search engine. I even know people who pay for DuckDuckGo.

Mike
And I'm
very sympathetic to that privacy angle, but I've always kind of reluctantly stuck with Chrome because it does integrate so well with all the other Google shit. I'm a persistent Android person. I have a Google Pixel phone. So that monopoly is convenient to you. The monopoly is convenient for me. my expectation, I guess, would be that if Yahoo buys Chrome, Chrome sort of immediately stops being as useful as it has been.

Erik
Oh, I thought you would be like, yes, I work for Yahoo. It's a fabulous, efficient company. If they bought Chrome, they could make it so much better. That was not my first thought, no. I think that's why people talked about those articles. It was like, Yahoo, honey Chrome, don't make me laugh. So the flip side of that is

Mike
that if they did spin off Chrome to Yahoo or to some other company, then I think that would be the motivation I need to switch to Firefox or to DuckDuckGo, which I think is probably a positive life change in general.

Erik
I just thought it was kind of sad that Yahoo is the punchline for this story. Like, oh, even Yahoo wants to buy Chrome. Apparently, the Yahoo search general manager, Brian Provost, he testified at this Remedies trial saying, yeah, we'd be up for buying it. And he also said that they have been, quote, actively internally developing a prototype of a browser at Yahoo to understand what it takes to put one out. And he said that Yahoo's in an ongoing discussion with other companies to talk about buying a browser, but he wouldn't say who. And then the article ended in a surprising way. they believe at Yahoo that buying Chrome would cost tens of billions of dollars. I was a bit shocked by that price tag, but I guess it makes sense. It's this dominant browser.

Erik
And it's
like hard to put a price tag on that, this thing they give away for free. But yeah, it's got this vertical integration with the search bar with ads and the search service and maps and everything else. So Yahoo's owner is a company called Apollo Global Management. and they apparently have enough money to spend tens of billions of dollars to buy Chrome. And at the end of the article, they say this. Apollo actually owns a browser brand that was at the center of a different antitrust case, but Provost said he wouldn't consider it an active browser. The brand's name? Netscape.

Mike
Yeah. You probably have to be an old person to get that. I used Netscape Navigator in the 90s. Many people did, but if you are a younger person and you haven't heard about it, there's good reason for that. It'd be rad,

Erik
though. If Yahoo buys Chrome, they rename it Netscape Navigator 3.0. That'd be beautiful.

Mike
I'm kind of confused by this story, to be perfectly honest, because I don't understand. And for one, how does Yahoo make money off this? Because I guess my assumption has always been that Google's motivation for building Chrome was to give them a platform where they have sort of absolute control over their ad platform. And clearly Yahoo does not want that. We think they would probably want to leverage their own ad platforms. Yeah. Anyway, I'm confused by the desire to own this particular software property.

Erik
The economics sound kind of weird, right? In the article, they write, Yahoo seems to agree that owning a web browser is a very important portal for search. The search general manager quoted above, Brian Provost, he testified that 60% of search queries are done through a web browser. Many people search directly from the address bar. That seems kind of weird. So we're going to spend tens of billions of dollars on this browser so that we can make it more likely that they'll use our search service. And then they will be more likely to see our ads. It economically sounds a little weird, doesn't it?

Mike
I totally am unclear on how they expect this to benefit them, especially if they spend tens of billions of dollars on it. But so far, nobody has asked me my opinion. Yeah, I mean,

Erik
here I am saying, oh, I'm surprised by this. I don't quite get it. Okay. It also feels pretty unlikely. So it felt a little bit like novelty news. I just brought it up because you used to work for Yahoo. And I thought, hey, if Yahoo buys Chrome, you can reach out to your old buddies at Yahoo and be like, hey, I'll come help you work on that browser. That sounds fun. Right, plug it forward in Rust. Great way to close out the career there. Yeah.

Mike
Yeah, get back into the big web company world. That would be fun.

Erik
Well, Mike, thanks for reading the news with me this week. A lot of interesting stories in the last couple of weeks. And I kept seeing articles like this Yahoo buying Chrome one and thinking, I got to tell Mike about this. But the one that lodged itself in my brain was definitely this comparison to the Luddites. I'd like to read more about the Luddites. There's some good books, recommended books out there. And there's some podcasts I'd like to listen to about the history of the Luddites to see what we can learn from that period. Friends, if you'd like to reach us, you can always send us an email, podcast.picturemecoding.com. We're also on Blue Sky at Picture Me Coding. And we also have Patreon sponsorships available for $4 a month. You can sponsor our show and help us host the show, help us pay for our hosting and other costs associated with the show, and continue to deliver the quality content you're hearing right now. Also, check us out on Blue Sky if you get a chance. All right, we'll see you again next week, friends. Thanks so much. Bye-bye.

Mike
See you next time.