It's Just A Model

012 - It's Just A Model - Unf_ck

• Peet Sneekes, Ron Kersic • Season 1 • Episode 12

Share episode

Use Left/Right to seek, Home/End to jump to start or end. Hold shift to jump forward or backward.

0:00 | 56:08

It's Just a Model is a biweekly conversation with Ron and Peet on the interstitium of innovation, tech, design, philosophy and behaviour.

Your hosts are Ron Kersic and Peet Sneekes.

In this podcast, we’ll be talking about the good, the bad and the interesting, all wrapped into an informal, unscripted conversation.

CHAPTERS

  • 00:00 — Intro
  • 04:25 — Destructive Design
  • 18:20 — "Citizens"
  • 34::32 — Aesthetics
  • 39:00 — Creative Destruction  
  • 40:48 — AI Manifestations
  • 45:05 — *Our* Skill Limitations 

LINKS

  • Frank Gehry — https://www.architecturaldigest.com/story/legendary-architect-frank-gehry-has-died-at-the-age-of-96
  • Shrek — https://www.ebritic.com/shrek-name-meaning/
  • Architectural Innovation — https://www.jstor.org/stable/2393549
  • Productivity Paradox — https://en.wikipedia.org/wiki/Productivity_paradox
  • Test-Driven Development — https://en.wikipedia.org/wiki/Test-driven_development
  • Domain-Driven Design — https://en.wikipedia.org/wiki/Domain-driven_design
  • Seeing Like a Desifn — https://en.wikipedia.org/wiki/Seeing_Like_a_State
  • The Jagged AI Frontier — https://mitsloan.mit.edu/ideas-made-to-matter/working-definitions/what-is-jagged-ai-frontier
  • Henry Rollins — https://en.wikipedia.org/wiki/Henry_Rollins
  • Henk Haaima — https://www.henkhaaima.nl/
  • Andrej Karpathy Interview — https://www.youtube.com/watch?v=kwSVtQ7dziU
  • AI Report Podcast — https://open.spotify.com/show/5Hpc8qDcawOEf4ulCouPau

ABOUT US

SUBSCRIBE

SPEAKER_02

Okay, Ron, are you ready? Very. Okay. Because this is It's Just a Model, a series on uh of ongoing conversations on the interstitium, very important word, of innovation, design, philosophy, and tech with my friend Ron Kursic and myself, Pate Snakers. You can find us on the internet. Please help us have jobs. I need a job. No, it's fine. Um, I just heard about an interesting uh assignment I just did, and I um so I'm kind of retrospectively, I'm known to make speculative design. I'm not sure if you're familiar. Yes, it's so it's designed for the people at home, it's designed for conversation. Basically, you show something and you don't tell anything to them, and you say, Okay, what do you think? And they say it's horrible, or it makes me think of murder, or I don't know, of plants. And uh that's basically the point of the design, so etc. So I make a lot of speculative design because I find interesting things interesting, and I like to shape it in things that are kind of like a little less unconventional. And uh, so I just gave uh got the feedback that the client said, Well, I'm not sure if I want to sign it. I made a poster with space for them to sign to put their autograph there, and uh they say, uh, well, I'm not sure if uh if I want to really want to sign this. So basically, we made them think about this possible future that they uh step into. So it was lovely. I was kind of proud of myself.

SPEAKER_00

It's a bit like uh the architect uh Frank Geary that uh he made Shrek models. Shrek models? Shrek.

SPEAKER_02

What is it it's like the Shrek the as in the the cartoon movie?

SPEAKER_00

Yeah, and the cartoon is made uh named after I think a Giddish term, Shrek. No, right? And you can probably also hear the German in there, Shrek.

SPEAKER_02

Yeah, yeah, yeah, yeah. It's like a monster kind of thing, or no?

SPEAKER_00

Oh and it's the Shrek Shrek in it uh in uh in in Dutch.

SPEAKER_01

Yeah.

SPEAKER_00

So he uh he was, he passed away recently. Famous architect, right? And typically when architects may make models, they're all fancy, you know, gleaming aluminum, and and he made them out of bin liners, and so that he so that you could have a conversation about the thing, the shape, yeah, yeah, yeah, its position, not the funkiness or the fanciness of the material.

SPEAKER_02

Yeah, the aesthetics should not uh uh be distracting, if you will. Yeah, yeah, yeah, yeah.

SPEAKER_00

Always thought that was very inspiring for what for whatever reason.

SPEAKER_02

It is, yeah. That's that's that's kind of like why I like making prototypes. Yeah. Because the it's obviously it's obvious that the medium shouldn't distract from the actual act. Yeah.

SPEAKER_00

And that's why I typically have very, I'd like to think, polished presentations.

SPEAKER_02

Oh, because it really needs to show this is finished. Yes. There's no in I'm not asking you, I'm telling you.

SPEAKER_00

And sometimes I come with just a big doodle.

SPEAKER_02

Yeah.

SPEAKER_00

Just lines and and and circles and and arrows pointing to things in space, etc. And it's the uh so last week I also did that, and a colleague of mine thought this was such a piece of rubbish that he just penciled over everything, and I thought that was actually my work. You're kidding, he said. That that was Brooke? I said, Yes, that was Brooke.

SPEAKER_02

But at the same time, that's awesome. That was my Shrek model. Yeah, yeah, yeah. Oh, that's interesting. There's a new word. Link in the show notes. Absolutely. By the way, did you know the show notes are as interesting, well, arbitrarily, as the conversation? So we like to think the conversation is interesting.

SPEAKER_00

I actually like to think we have this conversation so we can get to this cool list of show notes.

SPEAKER_02

Yeah, yeah, yeah. Yeah, really. So I put I put them in the in the various uh locations. And to me, it's like, oh, that's interesting. So I start reading, and all of a sudden then uh then the episode is late. Yeah, just there's nothing nothing to do about it. So uh I think uh we talked about it earlier in previous episodes. Uh, but I want to talk about um destructive design. I'm not sure. Did we talk about it, or was it just one of our notes?

SPEAKER_00

Probably or maybe not, but I love the term already.

SPEAKER_02

So, destructive design or destructive innovation is basically making things and introducing things new, and it haphazardly or intentionally destroys the old version of it. So, let's say if we have uh the loom, for example, it was a manual loom, and then uh we got uh machines like steam machines and motors, and they made things faster. And we also uh invented uh the punch card so we can program the loom, and it made like an entire industry of people working um uh obsolete, yeah. And so, and that's basically destructive innovation, yeah. It destroys like a big part of our society, a big part of our of an industry or an industry altogether, and it also it kind of disturbs me because and and we talked about the lexical gap uh in the previous episodes.

SPEAKER_00

That's when I can remember.

SPEAKER_02

Um, and the lexical gap is uh basically uh things that we don't have words for but are definitely definitely have changed. Uh and um just watch the previous episodes and it's pretty much well explained over there. Uh but the lexical I I feel like a lot of lexical gaps um appear in spaces where destructive innovation takes hold. And to me it's like okay, how can we make sure for how can we make sure the impact of destructive innovation isn't that big? Because sometimes we when we introduce a thing, it's not a success, or at least it is like a success, like a mass implementation, but we find out it's not uh good, it's not healthy, for example, for for a society. So one of the things is, for example, social media or mobile phones. Now we decided it's bad for kids, yeah, and um it will probably end up in being bad for everybody. You know, it's similar to smoking. Smoking is bad for kids, well, obviously. But to me, it's like okay, this is just one of the few steps that we take to make to say, okay, social media is inherently bad as as we know it right now, it's inherently bad for pretty much everybody. It's not for the weak-minded, like children, or like people that are uh less educated, it is bad for everybody because it uh taps into very base behaviors. And I found so what do you do when such a thing is introduced? What do you do if destructive innovation have has taken place and you basically want to undo this, right? You you basically want to apple Z this entire action, but we can't. So there are a couple of things that popped into my mind. One is software. When we build software, we basically build it for uh build it to fail. We have an extreme version of that, it's uh test-based uh uh development. It's basically you make all the tests that what might go wrong and first, and then you build the software. So when you uh when you put out the software, you basically say, okay, I know all the things that can go wrong, and I don't have red flags, therefore it is okay. Right? It's a very robust way to build it. And in software, we can also say, okay, this is a feature that basically didn't work, we roll it back, right? We undo the uh rollout and we replace it with something that was previously there. But we can't apparently we can't do that in real life. We can't Apple Apple Z social media on mobile phones. No, it's called path dependency, right? Right. Yeah. So it's uh that it's fascinating to me because I feel like we should, right? We should, with the rollout of social media, we should have had some way of hey, the a robust way. Well perhaps social media on mobile phones isn't the the best uh example. Let's say digitalization of everything, like uh bank services and like services of municipalities and stuff like that. It is nonsense to expect of people at this moment uh in time that people know digital design and ex and know where to go to expect certain things because digital on itself is fry uh is quite ephemeral, right? It's very when you shut off the screen, it's not there anymore. So it it lacks certain things, right? There's the lexical gap, it lacks the same fidelity and same uh addressability of a person sitting behind the desk that will help you figure things out. The the digit there is no digital equivalent for that. And um, and I feel like, well, let's say if you roll out such a surface as a municipality, like make everything digital, then you should have like redundancy or a way to roll it back or an exception or whatever.

SPEAKER_00

Well, I mean what you call destructive uh innovation design, yeah.

SPEAKER_02

Let's say destructive innovation because it's new, right? It's it's inherently with new things.

SPEAKER_00

Yeah, so what I as a uh part-time professor in innovation management had to say that uh would call is is uh system innovation. And you can see that I'm I'm a part-time professor in innovation management because I forgot the uh the authors of this term. Yeah the link will be in the show notes. It's it's on the tip of my tongue, literally. It's the uh it's the uh no, I mean so in I mean in innovation people often think that innovation is the thing, you know, there is it's is uh a new device or whatever. But yeah more often than not, the most impact in innovations are innovations in the systems surrounding the thing, yeah, right? And what you were talking about is in the end that you are affecting a system. Yes. Right? And uh and But that that's not what's being built. And that is the point I was getting to, right? I mean, the the reason why we you we get these effects is because we don't see it as an innovation of the system or a change in the system. We actually assume that the existing system will prevail and the and we force-feed or atom smash the new thing into the mold of the existing, yeah. Right? I mean it's the uh and and and and expecting it to act as similar, absolutely. And the whole digitization of uh municipality services is literally digitizing the existing system of service, yeah. Right? I mean if I I mean I I'm pretty sure this is correct, but if I go to uh the website of my municipality, on their website they actually have Lokatten. It's the it's the the things you can go to, right? It's literally the same word, yeah. Like the windows where you go to the windows where you expect like the lady or the so it's it's it's almost like a skewerphic uh it is it is right, and and then things go wrong because the new medium is not like that. No, and I think with social media the same thing, right? We we we we treat social media, I think, um, because I'm probably one of the few people that actually enjoys social media, and I have to say that as somebody who's glued to X for the whole weekend. It's amazing what's on there, really. It's the uh but but we treat it as a newspaper, right? Uh with content made by journalists that uh are uh are upheld back then to a journalistic standard. To be honest, not all journalists do uh anymore.

SPEAKER_02

No, yeah, no, also in the past it is a misnomer because everybody is slanted, and some are even like sponsored to by slanted uh and you see the same thing with uh with the with AI now, right?

SPEAKER_00

See, it took us what uh ten minutes something to get to get. No, but you see the same thing, right? We take generative AI and we automate certain things in organizations, and we never question if the organizational form is still actually suitable for generic AI. No, we just take the existing things, we automate what we can with generative AI, and what's left over is for people. Right. Right? The gaps in between is for people, and everybody's happy, and then we're all wondering why it actually doesn't bring any change, actually, or actually that the impact is a negative one.

SPEAKER_02

Well, if we measure, because that's the biggest mistake of all. Of course. We don't measure, we measure the thing, the new thing, but and and like all the things that it does, and if it does it well, yeah, but it doesn't measure the thing that was, and whether or not the same effect is being well, you can do a baseline, for example, but then you probably also do the baseline of the future service rather than the impact that it might have societally or whatever kind of thing.

SPEAKER_00

Or we take the measurements of the previous system and then we don't see any change. So we say this stuff is crap.

SPEAKER_01

Yeah, right.

SPEAKER_00

There is this very famous quote. I think it's from from Solo, Economist. Yeah, is that uh back then it was called IT. IT shows every shows up everywhere except in the productivity statistics.

SPEAKER_02

Yeah, yeah, yeah. Um that is a that's a brilliant quote. It's a brilliant quote. Because it's true, it is true. Uh I think productivity is uh both a misnomer, but also a nonsense uh uh goal for an organization to have.

SPEAKER_00

Ah, but that was the uh that was the the the the the yardstick for the corporation then, and and it still is for many corporations.

SPEAKER_02

Yeah, it's the old model, it's a factory model, it's literally the factory model, yeah.

SPEAKER_00

And then somehow IT doesn't show up in improving the factory model, or at need at least not at least not to the extent it was touted to. And you see exactly the same thing with generative AI. See, it doesn't do any productive improvement.

SPEAKER_02

Yeah, yeah, yeah, yeah.

SPEAKER_00

Yeah, maybe that's not the measurement.

SPEAKER_02

It's similar to uh the um the big AI chiefs, uh chiefs of staff. It's like they measure the amount of they used to measure the amount of code that you wrote as a coder. Yeah, it's like, well, that's odd. Yeah, that's odd. Because a really great coder arbitrary is is like a coder that makes as at least as least as possible code. Yeah, and over time even negative.

SPEAKER_00

Yeah, yeah, yeah. When I was a programmer. It's like delete the things. Yeah, when I was a programmer, not sure if I was a good one. But I did have a reputation that once I had my baseline that was working, then over time, uh, that means every weekend, the number of lines of the lines of code would decrease.

SPEAKER_02

Yeah, yeah, yeah.

SPEAKER_00

Yeah, it's which I thought was a good thing.

SPEAKER_02

Yeah, it's optimizing, right? It's it's it's making things charming, charming code.

SPEAKER_00

It is, yeah, it is. It's more readable and it's more out of your mode, and you can actually you can actually read my my my tests. So test-driven development, I have to correct your page. Oh, it's not just the things that can go wrong, it's also the things that should go right.

SPEAKER_02

Yeah, exactly. Yeah, yeah, yeah. I'm uh I stand correct or I sit corrected.

SPEAKER_00

Of course, I I always used to spend uh lots of time on my test up front. My test, so the names of my tests, when you read them, yeah should actually be the manual of the software.

SPEAKER_02

It is, yeah, yeah, yeah. And it is. Yeah. Yeah, that's that that's what it says in the manual of test-driven development.

SPEAKER_00

Yeah.

SPEAKER_02

It's like write it, write what you expect of the thing to do, and write and write what you do when things go wrong.

SPEAKER_00

Yeah. Up front.

SPEAKER_02

Yeah, exactly. And then write the code.

SPEAKER_00

Yeah, and if something goes wrong in your in your code that doesn't show up in the test, don't fix the bug. First create a test that should have flagged the bug.

SPEAKER_02

Exactly.

SPEAKER_00

And if you do that diligently and over time, you have this vast, vast suite of tests. I think if we Which make it very robust.

SPEAKER_02

If we if we start programming now uh with AI code with AI and code uh generators, we probably that will be the greatest a really great model. So for let's let's create a test language that which is very efficient and have that as an input.

SPEAKER_00

I have one agent that uh that writes the the tests for me now, and I have another one that writes the code that should pass those tests.

SPEAKER_02

Yeah.

SPEAKER_00

But they but they're separated. The one that gets to write the code doesn't get to write the test.

SPEAKER_02

Yeah, so one is the both ha of uh the both age of the other. Yeah, yeah.

SPEAKER_00

The bastard operator from hell, that's it, right? Something like that. I mean it's the so but no, but it it's really fascinating how old techniques come back in vogue again. Yeah, yeah, yeah. I mean, domain-driven design, right? I th I was quite I was both into it as an a bit skeptical about that one in 2006 because what design isn't domain-driven, okay. Right? It's the uh uh but that's now I mean it totally makes sense now uh uh when you have machines doing the well the the menial work, yeah, and we get to we get to think about the domain and how to specify it, yeah, and what's the language we use to describe the domain.

SPEAKER_02

Yeah, yeah, yeah. Yeah, that uh all of a sudden our soft language is can shape things um things in ways that we don't expect, right? So I I mean our natural language as we use it right now is very ambivalent. Uh to the contrary of code, which is a language that's quite exact, yeah, and that's a super and that's the reason why software projects fail. Yeah, yeah, yeah.

SPEAKER_00

Software projects fail because the subtlety of life falls through the crack, the crack of the formal language. Yeah. What programmers call an edge case is actually somebody's life, motherfucker. Yeah, I mean it always no that's an edge case. Or I mean, I I used to do a project for the tax office.

SPEAKER_01

Oh, yeah.

SPEAKER_00

And uh we found that uh in legislation there are two parts to get to an outcome, two different outcomes. And the uh the civil servant said, well, the citizen will complain if it is if it is wrong.

SPEAKER_02

But this is a very common way to think about it. Yeah. So bringing it back to destructive innovation, I find that our so our government in the Netherlands, they have this particular approach to pretty much all the things they introduce. So they have a project, they build it. So somebody has a great ID or Robert T N ID. They say, okay, let's make it a project. A budget is uh shapen, and some other party is building it, and they introduce it without checking. So there was this big upheaval of uh using um uh algorithms to select people that might be a little bit devious uh with the budgets that they got uh got from government, yeah. And uh so they wrote this algorithm, they basically made the list of people that uh might or might not be um uh devious, and they built them. They basically they they said, Okay, so we're giving you notes that you need to re uh reimburse the money that you got, or whatever kind of thing, or you even got uh penalty and stuff like that. And this was wrong because one the algorithm that wasn't tested at all, um uh spit out a list of uh people that were uh predominantly uh brown, and so there was this very racist skew on this particular algorithm, not intentionally, but it just was that was it that way. And rather than testing the algorithm first to see whether or not the list is actually uh okay or or uh valid, they just threw it out, and now they find out, now they need to deal with the ramifications, yeah. Which is interesting, right? It's an interesting approach to do things. It's I mean Uber basically uh used this particular uh algorithm. They just put it in a market, it destroyed pretty much all uh the entirety of taxi uh uh taxi life in many big um uh cities, and they let governments and the taxi uh people themselves solve the problem afterwards, solve solve the damage that has been done, and basically gobbled it up, right? But this is very costly if you address if you look at it on the individual uh uh basis. It is not acceptable when you see that an individual is all of all of a sudden without money, uh marriage broke, uh all kinds of personal things going wrong because they all of a sudden have a problem that they didn't uh create at scale because algorithm. And so government in this case said, okay, oops, we'll start thinking about making a solution. This algorithm was is basically a prime example of uh destructive innovation. It is something that's not tested enough or didn't have like the the the things, the the the cushioning around the project to make sure it actually uh ha would have had the effect that we intended to have.

SPEAKER_00

Yeah, I just would call it destructive, not even innovation.

SPEAKER_02

I mean it's the uh well it was something new right but rather than having people doing it they they basically ask asked an algorithm or created an algorithm to do the same job. Yeah. And indeed it is not innovation as we would say. No. But innovation for all intents and purposes doing something new with different means.

SPEAKER_00

Yeah I mean this this this reminds me of the book Seeing Like a State by George C.

SPEAKER_02

Scott I probably mentioned that there will be and there and a power so the reason by the way that we say link in the show notes is because we get a transcript and we just need to search for that particular word so we can put it in the show notes.

SPEAKER_01

Yeah.

SPEAKER_00

That they can see them that they can see them. Right? And that's why we we we we we we become numbers and we get numbers assigned to us and that's why we are citizens not people but citizens that need that need to fit into to the bigger scheme of government yeah and everything that doesn't fit into the bigger scheme of government is literally not seen doesn't exist.

SPEAKER_02

Yeah it's the machine eating the driver right it's it's so interesting.

SPEAKER_00

Yeah it's the uh and I yeah it is it yeah interesting is uh yeah is is the is the interesting word here.

SPEAKER_02

Well well we are I mean we could say there are now design methodologies that kind of pull back from that notion right it's it's a a machine should not create a new uh uh word list of how from their point of view the outside world looks like basically creating that uh lexical chasm right yeah like for because I I've worked with with with governmental people and they're lovely people but it seems like they their their perspective changes once they start working on the project and it's kind of intentionally it's like a cultural thing it seems like yeah um so to give you an example they start thorough talking about citizens rather than people yeah and to me that is which I found shocking by the way it is but it is shocking but it is also a coping mechanism it's a way for them to create a healthy distance between the things they do which affect the world significantly but also have like a common I don't know I think it's uh it's more of a commonality between the uh the the the governmental workers and the people they actually work for uh but it is something that need that they need to be aware of and I think they they are looking at the products that they make like for example the that implementation of the algorithm it is not sufficiently being um uh listened to right it's not they shouldn't talk about students or citizens or whatever kind of thing it's humans their the context of the people they influence is bigger than just yeah their banking account for example but that's for every system yeah digital or otherwise right yeah and that's the uh and and that's my um beef with this kind of thinking yeah all the details and nuance and nitty gritty of modern life or just life is is on seen is is made on seen. Exactly right and and if it wears its its hat it's called an ugly hat because it uh it it it it conflicts with our understanding of the world yeah and it needs to be eradicated right it's called a bug or an edge case or uh a one in a thousand that we can't be bother bothered with yeah it's actually reality it's uh it's a productivity statistic yeah that's it and that's why I love this new wave of technology that has this nitty-grittiness yeah yeah yeah so it basically it has a cushion if you will uh it can could so for let's say we have AI right we use AI for for systems to mold themselves to the lives of uh for example edge cases yeah so would you envision that the product would change itself for these particular individuals so so that the edge case basically doesn't exist the edge case is well handled rather than being the exception yeah oh yeah that it's uh that's uh that that's nice to say broad stroke right I think but it's the uh I mean at least the edge case is embedded in a technology that it's one big edge case yeah yeah yeah right I mean it's the uh and and we call it a jacket frontier right but basically that's just edge cases all over yeah yeah yeah it's great here and then one second it's really bad there I mean how how how can you it's like a whack a mole uh how how can you be PhD level and then nine year old um oh that's what I mean right I mean it's the uh it's it's and sorry there there's this there's this wonderful uh so do you know Henry Rollins of do I know Henry Rollins singer of Black Flag lovely yeah it's it's it's very it's very nice um so basically he he has this little accent kind of plays into what you say it's it pla it says it it it talks about uh he talks about his time in on the airport so he's basically standing in a row four people to go to board a plane and his observation was uh the that people in line with their ticket uh before they step into line they're like wonderful people socialize they're smart people etc and once they step into the line they're morons yeah and see it's life and and it's it's so it's such a poignant point to make because because it's because it's true right you see it but and the and the behavior that he spots is like the ticket and where you need to see see but because it's the seat is on your ticket and he observed like people walking into the aisle of the of the plane looking at the ticket looking at the chairs looking at the and it's like repetitively each and everything they start checking first the ticket and then the I mean you you will be able to remember like a number from one to fifty yeah right and uh but it's not it's like we we indeed change into morons in specific settings yeah and it's so it's so wonderful. Yeah but it is right it's it's it's it's mankind at mankind at least yeah probably life yeah it it I think it is life I but I and it kind of like can can we go to the to the next topic well but sure because um so I I read a uh a book a while ago yeah a book I don't really read that much uh apparently uh and it is the book about um the big Facebook uh kerfuffle that their data was being harvested by an extra Cambridge Analytica I think yeah and uh and this writer worked at Cambridge Analytica and he had this interesting cross interest of fashion politics and policy and that's a nice Van diagram it is it actually is so and to give you like the best expression of that is fascism so fascism has a dress code and to be if you if you regard it without all the all the bad aspects it's like it's pretty sharp right it's like everybody's basically well dressed and actually they put a lot of effort in it to create like this heroic uh suits for uh the power the the powerful people including uh the the army and stuff like that they basically exude power and also imply the role of the people that don't wear such things especially that yeah and um so he talks about it that it this is not an accident it is by by design and this is interesting to me because at this moment we have this very informal society we don't have dress coats per se i mean we do we have we have like niche cultures we have people with piercings and uh uh with a colored hair yeah we have people that can be considered punk but basically we have this melange of let's say meh yeah beige yeah beige yeah indeed beige is is the is the most acceptable uh acceptable color at this moment uh not per se the color but let's say the sentiment of the color and to me it's like okay so we are now entering in a society that is influenced by AI is there a way one for AI to to basically create like a an aesthetic in culture like our clothing the way our houses look like the way we interact with each other and stuff like that can we can we see like the telltale science already or how how might that look like I mean this is basically an open question because we can't answer it because we're in the middle of it right most of the time it's something that you can see retroactively or in a very abrupt way like for example fascism yeah but to me that's very interesting and to the other way around can we also design a culture and a society like the aesthetics but also the things that we do so we can create specific behavior from people and let's say flower power is was is a is an anti-direction is an anti-reaction to like the very formal uh 50s and 60s like productivity stay in line put your head down work hard etc and but flower power is what like the the reaction to to that and also expresses itself through like all kinds of things and I I'm still wondering it's like so can we shape that can we now introduce things into society that we can for example reintroduce color to also not also have our clothes be not beige but also our minds be less beige because we are very conformant we are very kind of like neutered in how we can deal with society you we can see that with all kinds of like um mental issues that we encounter these days more and more I think I'm not an expert and I'd like I'd like us I'd like a society to be resilient you know to be a little bit more independent I don't know I'm rambling at at this moment I I don't know I mean but if if I look at my Instagram feed yeah I see a whole stream of various aesthetics coming by yeah so it's the uh it's I'm I'm not gonna name them because I'm not gonna tell you the stuff I I I watch but oh yeah so we won't uh expose you guys no I mean something but it c it it covers the spectrum from uh let's call it uh Russian gold digger style all the way to um um nineties goth and anything I mean that's my biggest I mean that actually young people dress the way I dressed in the 80s.

SPEAKER_00

Yeah yeah yeah this is cyclical behavior yeah which is really amazing yeah and they copied it to a T. Yeah yeah yeah I mean just as a as an but now it but now it's considered like a well it was uh before of course but it's a uniform right it's it's similar it's similar to punk yeah well it's it's it's more of a signal than a uniform because um when I did it yeah there was one crucial sin and it is to copy someone.

SPEAKER_02

Yeah exactly so it's the You want to be a unique individual I'm unique so you dress like your peers I'm I'm unique no I'm not you know that's it's you're all individuals I'm not is that Monty Python it is Monty Python yeah you're all individuals yeah it's a it's one of my favorites yeah it's a really good one the Judean people front yeah oh you mean the people front of the Judea I want to be called Loretta okay I'm gonna stop now see I never can stop quoting Monty Python once I quoted Monty Python but it is just a very nice source of of culture of culture it's it it's it it basically well this is one of the main things right it openly criticized like things like religion which was very at that moment in time it's was quite risque yeah and uh but at the same time we lived in a society that this was not per se acceptable but it was uh welcomed yeah sufficiently so that the movie got made and nobody got killed yeah exactly yeah I mean yeah which you probably it it isn't a given that that wouldn't happen today no right I mean we have seen that like people making cartoons of certain uh icons yeah that uh may not be portrayed as icons yeah so yeah I mean yeah yeah back to a I I have no idea because I was thinking what what what did the internet do to aesthetics is there an internet aesthetic is there a worldwide web aesthetic if anything I would say that's the baseness of things because I'm always surprised that when that when I see people in China for instance young people is that they dress like young people here in the Netherlands.

SPEAKER_00

Oh how so will what would be an example literally the same I mean the same brands the same clothes the same trousers the same hoodies it's the globalized uh fashion it's the I mean I think I mean it it's I'm I'm just shocked how even niches yeah I mean that that I see Japanese people dressing as 80s uh London goths for instance yeah and to a T. And vice versa there is a certain also Japanese cyber god style that now everybody uh uh at least in my circles uh dresses up at when we go out and have fun. Okay it's the uh so it's but it's it's it's it's a niche but it's global. Yeah yeah yeah and I'm always really surprised how how how how uniform that looks like I mean and and didn't expect to see a shot from somewhere in China in one of the many large cities and literally see people dressed interesting as people dress when I go out of this building.

SPEAKER_02

Might that be just mass like the industrialization and also copy like let's say that we can make specific pieces of clothing for quite cheap yeah but also make it like a mass product pretty fast I mean we've seen in China that the factories that made like both clothing uh fashion things but also technology it's like when they're finished uh when they're out of contract with their uh customer they basically made the same product and use the same molds to make the fake thing yeah right yeah and it's like an optimize it's a way to optimize yeah in an industrial way so this is just industrial capitalism as it uh was meant to be okay that sound I wanted to sound profound but it's well it is profound but it's also kind of funny right it's it's like no we don't I mean we do it I mean it happened to be this case like we intended indeed to be things to be cheap and repetitive uh things to be uh um uh very customizable at uh uh at a certain uh point but it is it has negative effects it has like the digital same or digital sameness or just sameness yeah the the the beige of uh society maybe that's also just creative destruction right you yeah you create something and destroy something in order to create something you need to destroy something yeah if you make something new you lose something if you can see something new you can unsee some things you could see before yeah but that kind of leans more into the nature of things right it's like of course when you introduce something like for example if you buy a new sweater of course you will wear it more often than your old sweater right but it doesn't mean that your old sweater doesn't exist anymore and that's and that that is kind of like my point it's like the old sweater was also a fine sweater you you would you also want to Yeah but it went out of style because the new style made the old sweater look ridiculous yeah exactly so it doesn't exist but it isn't uh viable anymore you can literally not be seen in it anymore you want to unsee it yeah this is this is something we I I really need to think uh think about this a little bit more me too because this is uh to me this is very interesting so I really want to lean more into the design culture aspect of what technology and digital does yeah um funny enough like a couple of years like a relevant amount of years ago uh one of our former colleagues uh Hank Heima wrote like a small book or I think like a magazine kind of thing yeah yeah uh leaning into digital sameness yeah and addressing this yeah in particular.

SPEAKER_00

Yeah big shout out to Hank.

SPEAKER_02

Yeah definitely yeah Hank Heima is um well I mean we talk about it now and it kind of seems like a relevant topic at the moment still yeah or again especially now yeah yeah yeah yeah yeah so to me that's that's um yeah it's quite interesting.

SPEAKER_00

So you you also wanted to talk about uh our previous episode well yeah I mean two things right I mean in the previous episode I had this idea of writing an editor right yes I wrote the editor no no no I didn't write the editor I I I manifested the editor and something wrote it for me took a bit longer than I thought because the uh the Google Docs API is a bit more limited than I uh intuitively assumed oh but I now have this little markdown editor no yeah it's markdown editor that uh I write my things in there yeah and then in the evening AI comes by and start making sense of it and and and and reads it and then asks questions about it. So it doesn't say this is not clear or this is not that it actually asks why did you jump from A to X? And that's the uh and that's it's really fun. So it's so so nowadays in the more nowadays I mean it's just a week actually it's just this weekend. Yeah but this morning I woke up had my coffee at uh I tend to be an early riser so it was uh 5 30. So I had my first cup of coffee 5 30 and I open up my little app and I'm just wondering I mean yeah that that's a good thing why did I write this why didn't I imagine that well I mean let's literally sit in there and think yeah right and that's uh it's it's a really nice way of of reviewing not saying this is wrong or I don't understand why did you think this was understandable? It was less like a uh psychologist uh asking you yeah totally absolutely or or or your therapist uh meant yeah that that's the uh and that's what we talked about and you also wrote an article about uh uh in uh in response to episode 11 right yeah yeah which which is the cool thing by reading the article I actually understood what we discussed in the episode yeah I mean it's well it's my interpretation right yeah sure sure no but I mean it totally I mean I mean not to tote our our own own horn but the fact that we converge on the fact that generative FAI is not for automating the the the silly stuff and the stuff we don't like to do it's actually to help us do the stuff we want to do the creative stuff yeah and not only as a passive participant but as an active participant yeah yeah yeah because over the weekend I used um three different uh image models uh from uh OpenAI um Gemini and uh Grock to start creating the visuals for my presentations now visuals in my presentations are a thing to me yeah it is something you take pride in very yeah right and now I thought you know what I'm gonna try let's see if they it can make my visuals okay and here was the thing it's the uh so what I actually learned over by my the over this weekend probably about myself right is that you need to take yourself a bit out of the process you don't the more you specify very directly I want this make it yeah yeah it goes wrong yeah because it tries to do it and it doesn't it fails yeah so both your expectation becomes more precise yeah but also the the noise in the model yeah increases. Yeah and but you give it an intention you give it a context principles principles about Negative space, for instance, and the colors alike, and uh uh things to absolutely do, things to absolutely do not. Sit back and wait. And did it work?

SPEAKER_02

Yeah, fantastic. I'm really keen to finding out. Could you could you include some? I'm gonna include something on the same thing.

SPEAKER_00

Yeah, because I've been trying, because uh uh what I'm not very good at is making 3D uh images. Uh-huh. Uh I if I do, I do always the isometric things, yeah, yeah, yeah. Which I think are cute, but they're also easy to use. It's a great aesthetic. Yeah, but it's a simple algorithm, right? You just share and rotate and look, it looks the 3D. Yeah. That appeals to me. Uh-huh. But now I've got more proper uh 3D things that also animate, which are basically 10-second videos. But I make them look like an animation.

SPEAKER_02

So this kind of answers your transition, uh, your magical transition.

SPEAKER_00

Yeah, yeah, yeah. And and the the weirdly thing is, I mean, it's the uh people always bitch and moan that these things cannot do XYZ.

SPEAKER_01

Yeah.

SPEAKER_00

I actually don't uh think that these things are that limited in capabilities, but we are limited in skills, in enticing, unlocking capabilities. I mean, what I learned about myself is the skill to get through to these things.

SPEAKER_01

Yeah.

SPEAKER_00

It's the uh because I failed at uh Saturday morning and I was rather happy uh Sunday afternoon.

SPEAKER_02

Okay. So and who changed? You changed.

SPEAKER_00

I I changed.

SPEAKER_02

Yeah, yeah, yeah.

SPEAKER_00

I I I I learned the skill. No, I thought that this is this is really interesting. This is this is interesting. The limitation is my skill, and I hope everybody's skill. That is not just just me, and not their capabilities. So, do you think you might be uh a better?

SPEAKER_02

I mean, you you teach at this moment, right? This is your one of your main gigs. Yeah. So do you think you might be a better, let's say, student prompter by by doing this exercise?

SPEAKER_00

Yeah, it's the uh oh, that's that's a good one. I didn't uh think about this one.

SPEAKER_02

I mean it's like creating empathy in this case for an inanimate thing, but uh the basically the practice is the same.

SPEAKER_00

Yeah, I mean, so one of the things I uh I learned when I started teaching last year, so innovation management for the first time, right? Um I kind of do and get innovation. Yeah. So I thought, you know, I have all the materials. Yeah. So I thought easy peasy.

SPEAKER_01

Yeah, yeah, yeah.

SPEAKER_00

Until I start to write into syllabus. And then to actually write a syllabus, you know, there is a lot of white space that was just conveniently left blank, right?

SPEAKER_02

So I actually It's like the chapters with magic shit happens here. Yeah, yeah, it's the uh to-do or to be written, right?

SPEAKER_00

I mean, it's the or trivial, yeah, we'll do someday. Yeah. No, but I actually learned a whole bunch of stuff about innovation by teaching innovation.

SPEAKER_01

Yeah, yeah, yeah.

SPEAKER_00

It's really weird.

unknown

Right?

SPEAKER_00

And this is the same thing. I actually learned a lot about what I want to express by trying to get the machine to express it.

SPEAKER_02

Yeah, yeah, yeah.

SPEAKER_00

So I really think that I will be a better storyteller, communicator, student prompter.

SPEAKER_01

Yeah.

SPEAKER_00

Because I'm prompting these manifestations of people, students.

SPEAKER_01

Yeah.

SPEAKER_00

Right? So that it's the uh yeah, I mean, as you can probably see, I just I mean just flap a guess at what how fucking cool this technology is.

SPEAKER_02

It is, and then it was but also that it's untapped, right?

SPEAKER_00

It's very it's this huge misunderstanding of what this thing is. So this morning I saw an interview with Andre Capati. Um, he's one of the more well-known uh Link in show notes, but yeah. But he communicates well for an uh uh for an scientistslash tech person. No, he's really good at communicating.

SPEAKER_02

He's got credentials, but he also communicates well.

SPEAKER_00

Yes, yes. And he said, and uh uh link will be in show notes. Will also be in my Friday newsletter because it's that is that good I uh interview that um he was a bit of in a in an in an in a neurotic state because he couldn't stop coding nowadays. He was just coding for 16 hours. Ha! And I had the same thing, not coding but in visualizing. I actually woke up at 4 a.m. this morning, yeah, and I thought, you know what I'm gonna do? I sneak out and play with my little mates and make some yeah, it's that bad. No, this really remembers me when I got my first uh home computers. Yeah, yeah, yeah. Because I was slightly addicted. I also would sneak out and start programming at the home computer that was in our living room and turn the television uh the brightness down because otherwise my parents would see it. Yeah, it's the same thing. Yeah, or this maybe maybe this works. Oh, if I say this, maybe then it works. Oh, I should have done this. It's this whole unlocking things, yeah. And for 20 years I understood my medium, and the only limitation just I have to do it and have to press this, and now there's a new medium basically. That was a new medium, and and and I have to learn what I have to learn.

SPEAKER_02

Yeah. It's and this I I've heard this uh similar story with uh but he didn't reflect on it with uh Alexander Klipping of the AI report. Uh link in the show notes. I think it's it's kind of lovely. It it seems like to basically articulate the thoughts they already had, but nevertheless it's very interesting. But he also had this exact story. He's like, I sp I don't spend more time in programming like AI uh agents or whatever kind of thing he he's making. Uh but he did he did say he spent uh he enjoyed more of the time doing that. He basically was like you said, more enticed. Like, hey, if this is possible, then something else might also be possible, and then he starts starts tinkering basically.

SPEAKER_00

Yeah, indeed. I think it's the the the just the the the the the the time spent to to having fun ratio is massively higher. It is massively lower. Okay, yeah, yeah. Fun to time spent ratio. It's very fun, it's very fun, yeah. I mean, my uh I had a meeting that was 10 cut 10 minutes short today. The first thing I did is, yeah, don't laugh. It's that bad. The first thing I did is, oh, yeah, I have 10 minutes to play with Gemini now.

SPEAKER_02

So in the previous episode, I kind of like mused on that thought that this might be very addictive, similar to social media, similar to the. Okay, so let's let's we we got five more minutes. Let's let's tap into that. Therapy. Yeah, yeah. No, no, no, not therapy. I I I heard this wonderful thing. So I uh uh listened to this podcast, it's called No Agenda. It's like uh two people that are uh about uh 10 and 20 years older than I am, so they they lived life, yeah, but they're still living it, and they're also very much interested in AI and uh new media and stuff like that. And one of them has this policy, which is put the mobile phone in your drawer, so not in your underwear, but in the place where you put the stuff that you will never use, and um, that's basically how he does everything. And um, so they mused on an article or some text of somebody. I I I really don't know the source at the moment, but I'll try to figure it out. And this person says, We forget how to be bored, we are scared of being bored, and but the wonderful thing that happens when we are bored is that we start reflecting on life, we start having our own thoughts, we basically uh both defrag our thoughts, right? Reorder and try to recombine them in new ways and start to figure out things that kind of like are important in our lives, um, but also uh uh enjoying that. Because boredom isn't doing nothing and being bothered by it, but boredom is an activity that should be um is reinvigorating without us knowing. Yeah, right. We feel guilty if we feel bored, but boredom is important. So basically, he's the the thesis was I let's create boredom by leaving our phone home uh when we go out and have a walk. Yeah, uh not putting on headphones in particular. So I'm looking at the ladies, and I know it's a very protective way to feel like, but it's also kind of unhealthy because you shut yourself off for the outside world, and the outside world is in in some cases it's um it's important to follow along, right? Uh to uh to get more social cues and stuff like that. So to opening up to boredom and basically having your mind active in a different way that you do with consuming media, with interacting through uh social media for for example. Um we at as a as a nation kind of lost that and we need to reintroduce that because it will have ramifications if we don't. Yeah, yeah, I can see that. And that might also lean into because you you say it it's addictive, but I feel like it might be a it might be comfortable, right? More comfortable. It's it's a feeling that we know and it's a feeling that we we we what we crave at at so yeah that that that might that might be it's it's addictive.

SPEAKER_00

It's a happy form of uncomfortableness.

SPEAKER_02

Yeah, yeah, yeah. Yeah, in your case, right? Yeah, in my in my case, but it's like your type of person, yeah, probably, right? It's it's like you feel very comfortable in the unknown is in figuring things out that people might be scared to figure out, yeah. Yeah, but similarly, I mean there are lots of people that are comfortable in in other things.

SPEAKER_00

Yeah, I know, I know.

SPEAKER_02

They tell me, yeah, but uh so my thesis the the other episode was the reward that we get doing these particular behaviors is similar to doom scrolling, is similar to eating a lot of amounts of sugar, is similar to uh doing dangerous stuff for people that like to do dangerous stuff. Yeah, and um it we've we kind of like let that take and over take over. In uh, and so the things that we forgot, the lexical gap is that we forgot boredom and the very functional role it has in our lives. Yeah, so to me, it's like have a walk sometimes without any means of electronics. I will do this. So, this is my experiment.

SPEAKER_00

I'm gonna try this that this week too, and and see what it delivers. I'm gonna report out on uh episode episode 13. Seems like a very uh fitting episode.

SPEAKER_02

Oh, yeah, the doom. We need to do this on uh on a Friday somewhere. Let's see if we can sync up with the calendar. So on that note, I guess.

SPEAKER_00

Yeah, that's another hour.

SPEAKER_02

That's another hour. I really enjoyed this uh this one. I think I talked too much. So next time you need to prepare something, it's uh it's uh it's very nice, and I'm really looking forward to third team. Me too. And uh thank you so much for this lovely conversation. Thank you for me. Likewise. Right, see you later. Oh, by the way, uh make sure to visit uh it's just the model.com. You'll find information on Ron, but also the show notes, very important. And um, yeah, let's stay in touch. Next time. Ciao. Next time, bye bye.

Head Shot

Peet Sneekes

Host
Head Shot

Ron Kersic

Host