Brian Potter - Future of Construction, Ugly Modernism, & Environmental Review

1. 0:00 – 6:54

   Why Saudi Arabia’s Line is Insane, Unrealistic, and Never going to Exist

   1. DPDwarkesh Patel0:00

      Okay. Today, I have the pleasure of speaking with Brian Potter, who is an engineer and the author of the excellent Construction Physics blog, where he writes about how the construction industry works and why it has been slow to industrialize and innovate. And it's one of my favorite blogs on the internet. I highly, highly recommend that people check it out. Brian, my first question is about The Line project in Saudi Arabia.

   2. BPBrian Potter0:27

      (laughs)

   3. DPDwarkesh Patel0:27

      What are your opinions?

   4. BPBrian Potter0:28

      Okay, yeah. Just this will be a, a, a s- just in no particular order. Um, one, that it's just, you know, interesting that this country, Saudi Arabia, is in the Middle East, just kind of in general, is just so willing to just do these big, crazy ambitious building projects, uh, and just pour just huge amounts of money in, in constructing this, this infrastructure in a way that you don't see a huge amount in, like, the modern world, you know? They do it, China obviously does it, huge amounts. Um, there's some other minor places but in general, you don't see a whole lot of countries doing just these big, massive, uh, incredibly ambitious projects. So like, on that level, it's, like, interesting. It's like, yes, glad to see that you're doing this, but, like, the actual project is clearly insane and makes no sense.

   5. DPDwarkesh Patel1:21

      (laughs)

   6. BPBrian Potter1:21

      I mean, just from, like, a, a, just the physical arrangement layout, right? Like, there's a reason that cities, like, grow in, like, two dimensions, right? Like-

   7. DPDwarkesh Patel1:30

      (laughs)

   8. BPBrian Potter1:30

      ... a one-dimensional city is like the worst possible arrangement for just, like, transportation, right? It's like your maximum num- maximum amount of distance between any two points. And so, and y- you know, so just from that perspective, it's, like, clearly crazy. Uh, and there's n- there's no real benefit to it other than, you know, perhaps in some, like, weird hypothetical transportation s- situation where you had, like, really fast point to point transportation, uh, some sort of weird bullet train setup, maybe it would make sense. But there's no reason to build (laughs) there, in general, there's no reason to build, like, a city like that, right? Like even if you wanted to build, like, an entirely enclosed thing, which again, doesn't make a huge amount of sense, you would save so much material and effort if you just, you know, just make it a cube. Make it, I would be more interested in a cube-

   9. DPDwarkesh Patel2:20

      (laughs)

   10. BPBrian Potter2:21

       ... than, than The Line. But, uh, yeah, those are sort of my in- initial thoughts on it. I will, I'll be surprised if it, if it ever gets built, but-

   11. DPDwarkesh Patel2:28

       The cube, like, from the meme about, like, you could fit all the humans in a cube the size of Manhattan? Or-

   12. BPBrian Potter2:33

       Yeah, exa- Yeah or something like that.

   13. DPDwarkesh Patel2:35

       (laughs)

   14. BPBrian Potter2:35

       Or I mean, yeah, if you're just gonna build this g- big giant mega structure, at least take advantage of what that gets you, which is, like, you know, like, minimum surface area to volume, uh, ratio, stuff like that.

   15. DPDwarkesh Patel2:47

       And, and why is that important? Is that for, um, uh, temperature and stuff?

   16. BPBrian Potter2:50

       Well, basically the, it actually, I'm actually, this is actually interesting because I'm actually not how sure how sure it would work with a giant single city. In general, if, a lot of economies of scale come from, like, geometric effects where as something gets bigger, your volume increases a lot faster than your surface area does. So for s- like an enclosed thing, like a tank or a pipe or something like that, the cost goes down per thing of unit you're transporting because you can carry a larger amount-

   17. DPDwarkesh Patel3:19

       Mm.

   18. BPBrian Potter3:19

       ... for less, for a smaller amount of material. It applies to some extent with, like, buildings and construction because the exterior wall assembly is, like, a really burdensome and complicated and expensive assembly. So for, like, a really, a building with, like, a really big floor plate, for instance, uh, they can get, like, more area per unit, per amount of, like, exterior wall, like, all else being equal, be more efficient. I'm not sure how actually that works with, like, a single giant enclosed structure, because theoretically on, on a small level, it would apply the same way, like, because your, um, climate control is kind of a function of your exterior surface. Like, in, at some level, you get more efficient climate control, uh, if you have a larger vol- you know, larger volume and less area that it can escape from. But at a, for a giant city, I actually don't know if that works and it may be worse because you're generating so much heat that is now harder to pump out, like, d- dude, you know, the urban heat island effect and stuff like that where these cities generate, like, this massive amounts of, like, uh, just waste heat. I actually don't know if that would work if it, if it applies the same way. Um, I'm w- you know, I'm, there's like, I'm, I'm trying to reach back to, like, my physics classes in college and so I, I-

   19. DPDwarkesh Patel4:35

       (laughs)

   20. BPBrian Potter4:35

       ... don- yeah. Not sure of the actual mechanics of that but in general that's why, you know, why you would wanna perhaps build something of, of, of this size and shape.

   21. DPDwarkesh Patel4:45

       What, what was the thought process behind designing this thing? Because, like, Scott Alexander had a, a good blog post about, uh, The Line and where he said-

   22. BPBrian Potter4:52

       Mm-hmm.

   23. DPDwarkesh Patel4:52

       You know, a- presumably The Line is designed to take up less space and to use less fuel because you can just use the same transportation across the line. But, like, the only thing that Saudi Arabia has is space and fuel. So (laughs) like, how, what, what is the thought process behind this construction project?

   24. BPBrian Potter5:07

       (laughs) So I get the sense that a lot of Mideast, they're just, they're really, they've had some amount of success in building, like, big impressive physical construction projects that are just, like, an attraction just by virtue of, like, their size and impressiveness. Like, a huge amount of stuff, like, in Dubai, right, is, is, is something in, in, uh, this category. And they have that giant clock tower in Jeddah that is, like, similar, like, oh, biggest giant clock, you know, building, one of the biggest buildings in the world or whatever like that. So, like, I think on some level they s- think, they're expecting that you can just see a return just from building something that's, like, really impressive on s- and the biggest thing on some particular axis or whatever. So it could be, you know, there's some extent where I think they're just optimizing for, like, big and impressive and maybe not diving into it more than that. There's this, there's this theory that I think about every so often, it's called, like, the garbage can theory of organizational decision-making, which is that-... basically, the re- the, the choices that organizations make are not the result of any particular recent process. They are the result of whenever a problem comes up, they reach into the garbage can of, like, p- potentiatal solutions-

   25. DPDwarkesh Patel6:24

       (laughs)

   26. BPBrian Potter6:24

       ... and then whatever they pull out of the garbage can, that's what, uh, they, the decision that they (laughs) end up going with regardless-

   27. DPDwarkesh Patel6:30

       (laughs)

   28. BPBrian Potter6:30

       ... of how much sense that it makes. Uh, it was invented... It was a theory that was invented by academics to describe decision-making in academia. But I think about that, especially with, like, big bureaucracies and governments, I think about that a lot, where you can just imagine the horse-trading process and, like, how these decisions just evolve. You can just imagine that any random decision, especially when there's such a disconnect between, like, decision-makers and, like, the technical knowledge and stuff like that...

2. 6:54 – 10:10

   Designer Clothes & eBay Arbitrage Adventures

   1. DPDwarkesh Patel6:54

      Yeah. Um, tell me about your eBay arbitrage with, uh, designer clothes.

   2. BPBrian Potter7:00

      (laughs) Oh, man, you really did dive deep. Yeah. So this was a s- business, a small business that I ran for a few years, uh, s- seven or eight years ago at this point. Um, basically, a f- a hobby of mine was just high-end men's fashion for a while, which is a very strange hobby for an engineer to have, but, but there you go.

   3. DPDwarkesh Patel7:23

      (laughs)

   4. BPBrian Potter7:24

      Um, and basically, that hobby centers around, uh, for the enthusiasts of it, it kind of centers around, like, finding stuff for cheap because buying it new can just be, like, overwhelmingly expensive. But a lot of times, you can get it for very cheap if you're even a little bit motivated. It either shows up on eBay or it shows up in thrift stores if you know what to look for, um, especially because a lot of these clothes and things like that, they can last. Because they're well-made, they can last, like, a super, super, super long time. And so even if somebody who has worn it for 10 years or something, it might be fine. So a lot of this hobby centers on (laughs) finding, like, ways to get this, like, really nice stuff for, for cheap. Uh, and a lot of it is based around eBay. But it's, it was really tedious to find really nice stuff on eBay. Uh, you kind of had to, like, manually search, like, a bunch of different brands and then filter out obvious bad ones and search for, like, typos in brands that were, like, pretty reliably put in titles and stuff like that. And, um, so I was, I was in the process of doing this. Like, "Oh, this is really annoying. I should (laughs) figure out a way to sort of automate this process." And b- a- and so I made a very simple web app that basically you would search for, like, you know, shoes or something, and it would automatically search, like, the very nice brands of shoes, all the typos in the brand name, stuff like that, and it would just filter out all the junk and let you search through, like, the, the good stuff. And then I put a, you know, I set up a affiliate system basically, so anybody else that used it, they would, I would get, like, a kick of, of, uh, the, the sales. Um, and yeah, yeah, so while I was interested in that hobby, uh, (laughs) I, uh, I ran this website for a few years. And it was, it was, like, reasonably successful. It was one of the first things I did that, like, got it any real traction on the internet. But, like, it was never successful in proportion to how much effort that it took to, like, maintain it and, uh, update it and stuff like that. So kind of as I moved away from the hobby, I eventually stopped putting time and effort into maintaining the website. (laughs) I'm curious as to how you even dug that up.

   5. DPDwarkesh Patel9:28

      I have a friend who was with you at the, um, at the Oxford, uh-

   6. BPBrian Potter9:32

      Oh, gotcha.

   7. DPDwarkesh Patel9:33

      ... uh, the Refuges Conference, G- Conor Taberock.

   8. BPBrian Potter9:35

      Nice.

   9. DPDwarkesh Patel9:35

      I don't know if you remember him.

   10. BPBrian Potter9:36

       Oh, nice.

   11. DPDwarkesh Patel9:38

       Yeah, yeah.

   12. BPBrian Potter9:38

       Yep.

   13. DPDwarkesh Patel9:39

       Um, (laughs) no, uh, finding other information about you on the internet actually was quite difficult. Uh, you amazi- somehow managed to maintain your anonymity, so.

   14. BPBrian Potter9:46

       (laughs)

   15. DPDwarkesh Patel9:46

       And then what were, i- if you're willing to reveal, what was, like, the PnL o- of this project?

   16. BPBrian Potter9:50

       Oh, it made, uh, maybe a few hundred dollars a month for a few years. But I only ever ran it, like, as a side hobby business basically.

   17. DPDwarkesh Patel10:00

       Mm-hmm. Gotcha.

   18. BPBrian Potter10:00

       So in terms of, like, ti- effort, you know, per my effort or whatever, I'm sure it was, like, very low, you know, just, you know-

   19. DPDwarkesh Patel10:06

       Yeah.

   20. BPBrian Potter10:06

       ... pennies an hour or something like that.

   21. DPDwarkesh Patel10:08

       (laughs)

3. 10:10 – 19:28

   Unique Woes of The Construction Industry

   1. DPDwarkesh Patel10:10

      So a broad theme that I've gotten from your post is that the construction industry is plagued with, um, these lossy feedback loops, um, you know, a lack of, um, uh, y- uh, strong economies of scale, uh, regulation, and then, you know, cost, um, uh, mistakes being very costly. Do you think that this is a general characteristic of many industries in our world today, or is there something unique about construction?

   2. BPBrian Potter10:36

      Interesting question. I mean, that's kind of one thing when you think of it is there's a lot of, like, individual factors that are not unique at all, right? There's, like, you know, construction's highly regulated, but it's not necessarily more regulated than, like, medical devices or, like, jet travel or even probably cars to some extent, uh, which are very, you know, have a whole battery of, of, of performance criteria that they need to hit. So and like, you know, similar with, like, coupled to, like, land use or something like that, people say like, "Oh, it's, you know, the, the land requirements and have to build it on site, uh, make it difficult." But there's a lot of things that fall into that category that, again, don't really share the same, you know, structure of how the industry works than, than construction does. So kind of it's the, um, I think it's just kind of the interaction of all those effects. One thing that I think is perhaps underappreciated is that a building is, like, really... And then, again, this is not unique either, but a building is really highly coupled in a way that, um, uh, the systems of a building are really highly coupled in a way that a lot of other things aren't. Um, if you, you know, if you're manufacturing, like, a, a computer or something like that, right, the hard drive is somewhat independent from the display. It's somewhat independent from the power supply. I mean, these things are coupled, but they can be built by independent people who don't really necessarily even talk to each other and assembled into one structured thing. A building is not really like that at all. Um, every single part affects every single other part for the most... In some ways, it's, you know, it's, uh, it's, it's, it's like biology. And so it's very hard to change something that doesn't end up disrupting something else. And part of that is because the job that a building does, which is to create...... a controlled interior environment. You know, basically every sil- single system has to run through and around this, this sort of surfaces that are creating that controlled interior. So everything is, like, kind of touching each other. Um, and again, that's not unique of, you know, a plane or anything really highly engineered. An iPhone kind of shares that too, to some extent. But the, the, in terms of, like, the size of it and the relative small amount that you're paying in terms of, like, unit size or unit mass or something, it's quite low.

   3. DPDwarkesh Patel12:50

      Is the fundamental reason you can't, like, have as much specialization and modularity, is that, is that basically tre- uh, transportation cost?

   4. BPBrian Potter12:57

      Yeah, I think it's really more just the way a building is, right? Like, you can't, you know, for the, for example, for the electrical system of your house, you can't have, like, a separate box that is your electrical system that you would just install and if you needed to replace-

   5. DPDwarkesh Patel13:09

      Mm-hmm.

   6. BPBrian Potter13:09

      ... the electrical system-

   7. DPDwarkesh Patel13:10

      Gotcha.

   8. BPBrian Potter13:10

      ... you could take the whole box out and put the new box in. The electrical system runs through the entire house. Same with plumbing, same with, like, the insulation, same with, you know, the interior finishes and stuff like that. There's not, like, there's not a lot of, like, modularity in, like, a physical sense.

   9. DPDwarkesh Patel13:28

      Gotcha. Okay. Now, uh, Ben Kuhn had this interesting comment on your article about, um, h- why it's hard to innovate in construction, where he pointed out that many of the reasons you give for why it's hard to innovate in construction, um, like the sequential dependencies and the highly variable, like, delivery timelines, that that's also common in software where, uh, Ben Kuhn works. So, um, why, why do you think that the same sort of, uh, stagnation has not hit other industries that have a s- superficially similar characteristics like software?

   10. BPBrian Potter14:02

       You know, what, how I kind of think about that is that you s- you kind of see a similar structure in anything that's kind of, like, project-based or anything where you're kind of having to, like, there's an element of figuring out what you're doing while you're doing it, you know, compared to, like, a large scale manufacturing option, right? You spend a lot of time figuring out what exactly it is that you're building, and then you spend a lot of time designing it to be built. And then, you know, your first number of runs through it, right? You tweak your process to make it more efficient. But at some point, you... A- and there's always an element of, like, tweaking it to, to make it better. But to some extent, the spec process of, like, figuring out what you're doing is largely separate from, like, the actual doing of it yourself. For a lar- for a project-based industry, it's not quite like that. You're kind of having to build your process on the fly while you do it. And of course, there's best practices that shape it, right? Like, for somebody writing a new software project or anything other project-based, right? Like making a movie or something like that, they know roughly how it's going to go together but, you know, there's going to be a lot of unforeseen things that kind of come up like that. I think the biggest difference is that either the pr- th- those things can often scale in a way that you can't with a building. Like once, if you, once you're done with a software project, you can deploy it to 1,000 or 100,000 or a million people, right? Once you're d- once you, um, once you finish making a movie, again you can, or 100 million people can, can do it or whatever. It doesn't quite look the, uh, the same with, uh, with a, a building. You don't really have the ability to, like, spend a lot of time (laughs) up front figuring out how this thing needs to go. You kind of need to figure out a way to get this thing to go together without spending the huge amount of time that maybe would be justified by the, the size of it. And yeah, he, he's definitely right. I mean, I, I, I was able to dig up a few references for, yeah, software projects and how often they just have these big log pails of sometimes they just go massively, massively over budget. A lot of times they just, you know, don't get completed at all or, you know, but if, which is shocking. But the e- because of how, how many people ca- it can then be deployed to after it's done, the economics of it are, are slightly different.

   11. DPDwarkesh Patel16:19

       I see. Yeah. There's a fame in, uh, famous law in software, I forget the name of it, which is that a project will take longer than you expect even after you account for the fact that it will take longer than you expect.

   12. BPBrian Potter16:29

       Yeah. Hofstetter's Law or something like that is what I think it is.

   13. DPDwarkesh Patel16:33

       Yeah. But so I'm curious what the, what the lack of skill in construction implies for startups? Like famously in software, the fact that there's zero marginal cost to scaling to the next customer, that's like a huge boon to a startup, right? But the, the entire point of which is just scaling with like a hockey, uh, exponentially. Does, does that like fundamentally, uh, constrain like the size and quantity of startups you can have in construction if the, if the same scaling is not available?

   14. BPBrian Potter17:00

       Uh, yeah, that's a really good question. I mean, I, I, the obvious first part of the answer is that for software, obviously if you have a se- construction software company, you can scale it just like any other software business, right? Um, I think for, like, physical things is a lot more difficult. And yet, this, like, lack of, like, zero marginal costs tended to bite a lot of startups, right? Not just construction ones. But yeah, it's definitely a thing. Construction is, I think, particularly brutal because the just margins are so low and then just the empirical fact that trying to mo- what seems like it would be a more efficient method of building doesn't actually let you, uh, do it cheaper and like get better margins. Like the startup that I used to work at, Katerra, their whole business model was basically predicated on that. You know, "Oh, we'll just build all our buildings in these big factories and get huge economies of scale and reduce our costs and then make the billions of dollars that we're pumping into this industry, uh, or to, to this business, we'll be able to recoup it." And it, you know, the, the math just does not work out. You can't build... In general, you can't build cheap enough to kind of recoup those giant upfront costs. A lot of, a lot of businesses have kind of been, been burned that way and the, the most success you see-... and, like, prefab type stuff is, like, on the higher end of things where you can get higher margins. And so a lot of these, these companies, these, like, prefab companies and stuff like that, they tend to target the higher end of the market, where there is a... and you kind of see a few different premiums for that. Um, you know, the obvious one is that if you're targeting the higher end, you maybe have higher margins. Uh, and if you're building to a higher level of quality, that's easier to do in a factory environment. And so you're, you end up being, like, cheaper than, or at least com- a lot d- the delta is a lot different, less enormous than it would be. You know, you can build it a little bit, uh, building a high-level quality is easier to do, uh, in a factory than it is in a, uh, in the field. Uh, there's a lot of buildings that are like, or houses that are, like, built to a really high level of, like, uh, energy performance. For instance, like, need a really, really high level of air sealing to, to minimize, like, how much energy this house uses. You tend to see a lot more houses like that built out of prefab construction and other, like, factory built methods because it's just physically more difficult to, uh, achieve that, uh, on site.

   15. DPDwarkesh Patel19:27

       Can you say

4. 19:28 – 26:27

   The Problems of Prefabrication

   1. DPDwarkesh Patel19:28

      more about why, uh, you can't use prefabrication in a factory to get economies of scale? Um, is it just that the transportation cost will eat away any gains you get or is there, like, what is, what is going on?

   2. BPBrian Potter19:37

      Yeah. There's a, there's a combination of effects. And I haven't worked through all this. This is gonna be the next blog post basically, so I'll-

   3. DPDwarkesh Patel19:43

      Oh, excellent.

   4. BPBrian Potter19:43

      ... fig- uh, figure it out more by then. Um, but the very basic hi- at a high level is that, yeah, basically your savings are not, that you get from, like, using less labor or whatever, like that are not quite, uh, enough to offset your, your increased transportation cost. One thing about construction, especially like, like single family home construction, is that a huge percentage of your costs are just the materials that you're using, right? Like, a single family home is roughly 50% labor, 50% materials for the construction costs. And then you have development costs and stuff like that, land costs, things like that. And so, um, a big chunk of that you just can't move to the factory at all, right? Like the foundation, you can't, it's, you can't really do that in a factory. You could prefab the foundation but it doesn't gain you anything. Your excavation still has to be done on site obviously. So a big chunk of that can't move to the factory at all. And for ones that can, you still basically have to pay the same amount for, uh, materials. If you were doing it, like, theoretically that wouldn't be true if you were doing, like, hu- truly huge volume where you could get material volume discounts, but even then it's, you're probably not looking at, like, massive savings. So you can cut out, like, a big chunk of your labor and you do see that in these, in, like, factory built construction, right? Uh, like these prefab companies or, like, mobile home companies, they have, like, a really small fraction of, uh, of labor as their cost which is pretty typical from a factory in general. But then they take out all that labor cost, they still have their high material cost, and then they have this overhead cost of, like, whatever the factory is costing them, right? And then you add your additional overhead cost of just, yeah, transporting it to site which is pretty limited. And just the math (laughs) does not really work out in favor of the, um, in favor of, of prefab in terms of, like, being able to build, like, dramatically cheaper, right? You can obviously, you can obviously build a, uh, a building in a, as a prefab, using prefabricated methods and build a successful construction business, right? That, like, many people do. Uh, but in terms of, like, dramatically lowering your cost, you don't really see that.

   5. DPDwarkesh Patel21:50

      Yeah. Yeah. Austin Vernon has an interesting blog post about this, why there's not more prefabricated homes. And then d- the two things he points out were, like, transa- uh, transportation cost and the other one was that people prefer to have homes that have, like, unique designs or unique features. When I was reading it, uh, it actually occurred to me that maybe they're actually both the result of the same phenomenon. So, like, um, have, have you heard of, like, um, I don't know if I'm pronouncing it correctly, but the Alchian Alan theorem in economics?

   6. BPBrian Potter22:16

      Uh, maybe. I don't think so. (laughs)

   7. DPDwarkesh Patel22:18

      Basically it's the idea that if you increase, uh, the cost of, uh, some, uh, like, category of goods in a fixed way, like let's say you tax oranges, right? Um, uh, like added $1 tax to all oranges or I don't know, if transportation for oranges gets $1 more expensive for all oranges then people will shift consumption towards the higher grade variety because now the ratio of the cost between, like, the higher-

   8. BPBrian Potter22:44

      Mm-hmm.

   9. DPDwarkesh Patel22:44

      ... the more expensive orange and the less expensive orange, that ratio has decreased because of the increase in fixed cost. And, um, it- it seems like you could use that, uh, that argument to also explain why people have, like, strong preferences for uniqueness and all kinds of design stuff in manufactured houses which is that-

   10. BPBrian Potter23:02

       Mm-hmm.

   11. DPDwarkesh Patel23:02

       ... since transportation costs are so high, that's, like, basically a fixed cost and that that fixed cost has the effect of making people shift consumption towards, like, higher grade options.

   12. BPBrian Potter23:09

       I definitely think that's true. I basically, I would maybe phrase this as the construction industry makes it relatively, comparatively cheap to, like, deliver a highly customized option compared to, like, a really repetitive option, right? And so people... yeah, it's the ratio between a highly customized one and just, uh, you know, a commodity one is relatively small and so you see kind of the industry built around delivering somewhat more customized options. I do think that the... this is, you know, this is a pretty broad intuition that the, um, that people just desire too much customation- customization from their homes and that really prevents you having a mass produced offering. I do think that that is true to some extent. I mean, to one, I think one example is this, like, the Levittown houses which were originally built just, you know, in huge numbers of, like, exactly the same model over and over again and eventually they kind of had to change their business model to be able to deliver more customized options because the market shifted. I do think that the effect of that is basically pretty overstated. Just empirically you see that-... in practice, home builders and developers will deliver fairly repetitive (laughs) ho- housing. Uh, they don't seem to have a really hard time doing that. As an example, I'm living in a new develop- housing development that is just, like, three or four different houses copy/pasted over and over again, a group of 50. Uh, the developer is building a whole bunch of other developments that are very similar in this area. My in-laws live in a very similar development in a whole different state. Um, and if you just look at, like, multifamily or apartment housing, right? It's like identical apartments, you know, copy/pasted over and over again in the same building. There are a bunch of different buildings in the same development, right? It's not... You're not seeing, like, huge amounts of uu- uniqueness in these things. People are clearly willing to, uh, to just live in these, like, basically copy/pasted apartments. And it's, it's al- it's also quite possible to get a pretty high amount of product variety using a relatively small number of factors that you, that you vary, right? I mean, the car industry is like this, right? Where you can... There's enough customization options where... (sighs) I was read- I was reading this book a while ago that was basically pushing back against the idea that the car industry, uh, didn't... You know, the, the, the pre- you know, in the '50s and '60s was just offering a very uniform product. And they did basically the math and the number of customization options on their car were like more than the atoms in the universe. Basically, just there's so many different options to, like, all the different-

   13. DPDwarkesh Patel25:48

       (laughs)

   14. BPBrian Potter25:48

       ... permutations. You know, leather seats and, you know, this type of stereo and this type of engine. If you add it all up, there's just, like, just a huge massive number of, um, different combinations. And yeah, and just you can obviously customize a house just a huge amount just by, like, the appliances that you have and the finishes that are in there and the paint colors that you choose and the fixtures, uh, and stuff like that, that would not really change, theoretically, the underlying way the building goes together. So I do think that that idea, that the sort of fundamental demand for variety I- is a major obstruction, I feel like I, I do not think there's a whole lot of evidence for that, um, in the construction industry.

5. 26:27 – 32:20

   If Building Regulations didn’t exist…

   1. DPDwarkesh Patel26:27

      Interesting. Now, so I asked on Twitter what I should ask you, and usually I don't get, uh, interesting responses, uh, from this.

   2. BPBrian Potter26:33

      (laughs)

   3. DPDwarkesh Patel26:33

      But the, the quality of the, uh, th- the people in the audience that knows who you are was so high that actually all the questions I got were really interesting, so I'm just gonna ask you some questions from Twitter.

   4. BPBrian Potter26:44

      Okay. (laughs)

   5. DPDwarkesh Patel26:45

      Okay, so Connor Tabrock asks, "What is the most unique thing that would or should get built in the absence of construction regulation?"

   6. BPBrian Potter26:53

      Yeah, interesting. Unique thing. Yeah, unique is an interesting qualifier. I mean, there's a lot of things that just, like, should get built, right? Just, like, huge more house- huge amounts of more, uh, additional housing or just, like, creating more land where in these sort of really dense urban environments, where we need it. Like, you know, places like San Francisco, just fill in a big chunk of that bay. It's basically just mudflat, and put more housing on it. I saw a proposal like that earlier. Um, unique thing is, is, is, um, more tricky. One idea that I, that I really like, I read this in, um, the bo- the book Where is, Where is My Flying Car? But that it's basically crazy that our cities are designed with roads that, like, all intersect with each other. It just imme- you know, that's, this is a crazy, um, way to, like, structure a material flow problem, basically. And like any sane city would a- would be built with, like, multiple layers of, like, transportation, where each one went in, like, a different direction. So your flows would just be massively, massively improved. Yeah, so that just seems like a very obvious one, that if you were building your cities... That's not... That's slightly off-topic of the question, but if you were building your cities from, from scratch, right? Had your druthers, you would clearly want to build them in, like... And, and knew how big it was gonna get, right? So you could plan lo- very long term in a way that, you know, again, almost no cities did, but be able to plan so these, these transportation systems don't (laughs) , don't intersect with each other. And you have the space to scale them or run as much throughput through them as you need without, like, bringing the whole system to, to a halt. Um, there's a lot of evidence that basically says that cities, you know, tend to scale based on how much you can move from point A to point B through them, and I do wonder if you, like, you change the way that they went together, you could a lo- unlock, like, massively different cities. Or, or, or, or ev- or even if didn't unlock massive difference, but changed, like, the agglomeration effects that you see in, in cities if people could move from point A to p- point B, uh, much quicker than they can.

   7. DPDwarkesh Patel28:49

      Yeah, I, uh, I did an episode about the book, uh, Where's My Flying Car? with, um, with, uh, Rohit, uh, Krishnan, and I don't know if we discussed this. But yeah, th- that was an interesting part of the book, where he talks about, like, uh, uh, transistor design. Like, if you design transistors this way-

   8. BPBrian Potter29:04

      Yeah.

   9. DPDwarkesh Patel29:04

      ... like can you imagine how slow they would be?

   10. BPBrian Potter29:05

       (laughs)

   11. DPDwarkesh Patel29:07

       Okay, so, uh, Simon Grimm asks, "What countries are the best at building things?"

   12. BPBrian Potter29:13

       Uh, yeah, this is, uh, a, a good question. I'm gonna, again, sort of cheat a little bit and do it in space and in, and in time. Because I think a lot of the most countries that are most interesting and doing a good job at building massive amounts of stuff are not ones that are basically doing it currently. I mean, the current answer is, is, is like China, right? Where they're just building, you know-

   13. DPDwarkesh Patel29:36

       Yeah.

   14. BPBrian Potter29:36

       ... you know, more, more concrete used in the last 20 years or whatever than the entire world (laughs) used in, in the time before that, right? Just, like, big, massive amounts of urbanization. A lot of, like, really interesting buildings and, and construction. I mean, just in terms of, like, raw output, right? You, you gotta give it to them. I would also put, like, Japan in, like, the late 20th century on, on there. Um, basically at the peak of, like, concern. Like, wonder, like, is Japan gonna, gonna take over the world? And they, you know, they were really interested in building their, their stuff, um, quite quickly. They spent a lot of time and effort trying to use their, like, robotics expertise to try to figure out how to build buildings, uh, a lot more quickly. Um, so they had these, like, really...... interest in factories that were designed to basically extrude an entire skyscraper just going up vertically. Um, and, and all these, like, big giant companies developed them. So they had, like, these many, many different factors that were kind of trying to do this with all these robotics. Really interesting system that did not end up ever making economic sense but, uh, it is, it is very cool. Um, the, uh, the tr- the... I think they're sort of big, um, industrial, um, policy, um, organs of the government basically encouraged a lot of these industrial companies and stuff to, like, basically develop prefabricated housing systems. And so you see a lot of really interesting systems, uh, developed from these, uh, from these, um, sort of industrial companies in a way that you don't see in a lot of other places. The US basically just from, eh, 1850 to maybe 1970, so for like 100 years or something, um, just building, like, just huge massive amounts of stuff and in a way that, like, lifted up, like, huge parts of the economy, right? Like, you know, they, we spent all this, you know... I don't know how many thousands of miles of railroad track that the US built between like 1850 and 1900, but it was many, many, many thousands of, of, of thousands of miles of it. And of course, needing to lay all this track and build all these locomotives really sort of forced the development of the machine tool industry which then led to the development of, like, better manufacturing methods and inter- interchangeable parts which of course then led to the development of the auto-industry and of course that explosion just led to even more big giant construction projects. Um, so you really see that these just sort of being able to build just big massive amounts of stuff really in this virtuous cycle with the US really advanced a lot of technology and really sort of raised the standard of development just for a s- for a super long period of time. Um, yeah. So those are kind of my, my three answers.

   15. DPDwarkesh Patel32:19

       Yeah.

6. 32:20 – 44:45

   China’s Real Estate Bubble, Unbound Technocrats, & Japan

   1. DPDwarkesh Patel32:20

      Tho- those three bring up, uh, three additional questions.

   2. BPBrian Potter32:22

      (laughs)

   3. DPDwarkesh Patel32:22

      One for each of them. Um-

   4. BPBrian Potter32:24

      Shh.

   5. DPDwarkesh Patel32:24

      Tha- uh, that, that's really interesting. Um, ha- have you read The Powerbroker? The, the, the book about Robert Moses?

   6. BPBrian Potter32:29

      I think I got like a tent- like a tenth of the way through it. Uh.

   7. DPDwarkesh Patel32:33

      (laughs)

   8. BPBrian Potter32:33

      Sorry if that's-

   9. DPDwarkesh Patel32:34

      Tha- that's basically a whole book in itself, a tenth of the way. (laughs)

   10. BPBrian Potter32:36

       Sorry... Yeah, sorry Robert Caro fans.

   11. DPDwarkesh Patel32:38

       No, but wha- I, I'm, like, uh, half the way through it and so far it's, like, basically the story of how this one guy built a startup within the New York State Government that was just so much more effective, uh, at building things, didn't have the same corruption and clientelism and incompetence of the Tammany Hall political machine at the time. And, you know, maybe, maybe it, like, turns into tragedy in the second half, but so far it's like, w- we need this guy, like, wh- where do we get a second Robber- Robert Moses? Do you think that if you had more people like that in government or i- i- in the construction industries public works would be, um, um, more effectively built or i- is that just, um... I- is the stagnation there just a result of, like, o- other bigger factors?

   12. BPBrian Potter33:20

       Yeah. That's... It's an interesting question. Yeah. I, I, I remember reading this, this article a while ago that was complaining about, like, how horrible Penn Station is in New York and they're basically saying like, "Yeah. We basically... It would be nice to return to the era of, like, the sort of unbound technocrat where, like, these peop- these technical experts in, like, these high positions of power in government could essentially do kind of whatever they wanted to some extent, right?" And if they thought something should be built somewhere, they basically had the power to do it. Uh, you know, it's just kind of a facet of this problem of just how it's really, really hard to get stuff t- built in the, in the US currently. And I'm sure part of it is just that you don't see, like, these really, you know, talented technocrats or whatever occupying these, these high positions of government where they can, where they can get stuff done. Uh, but it's, it's, it's not super obvious to me whether that's, like, the limiting factor, right? I kind of get the sense that you... They would end up being bottlenecked by some other part of the process and just, like, the whole sort of interlocking set of institutions has just become so risk-averse that they would end up just being blocked by s- ... You know, in a way that they wouldn't if, you know, when they were operating in, in the 1950s and '60s or, or whatever.

   13. DPDwarkesh Patel34:43

       Yeah. Yeah. That's interesting. Okay. So speaking of Japan, uh, I, I, I just recently learned that they have... Uh, like a lot of the construction there is that they just keep tearing stuff down every, every 30 to 40 years and rebuilding it. So you have this interesting series of posts on how you'd go about building a house or a building that lasts for 1,000 years. But I'm curious, how would you build a house or a building that only, uh, lasts for 30 or 40 years if you're building in Japan and you know they're gonna tear it down soon? What, what changes about the construction process?

   14. BPBrian Potter35:12

       Uh, yeah. That's interesting. I mean, I've... As... I'm not an expert on Japanese construction, but I think, like, a lot of their interior walls are, are basically just, you know, paper and, and stuff like that. Um, I actually think it's, it's kind of surprising they use a somewhat for their... A lot of their homes, last time I looked, they used like a sort of surprising post and beam construction method which is actually somewhat labor-intensive to, to do. The US back in, like, the early 1800s used a pretty similar method and then once we started mass producing dimensional lumber, we kind of (laughs) stopped doing that because it was much cheaper to build out of two-by-fours than it was, like, big heavy posts and feet and mortice and mortise and tenon connections, stuff like that. I mean, I think the boring answer to that question is that you would just build like we build mobile homes essentially which is just using pretty, you know, pretty thin walls, pretty low-end, uh, materials that are just goned together pretty, you know, in a, in a pretty sort of, you know, minimal way which ends up being not that different than the actual construction method that single family homes use but just even further economizes and tightens the use of materials, you know, where a single family home might use, uh, a half inch plywood or whatever, they might try to use, you know-... three-sixteenths or, or even, or an eighth-inch plywood or, or something like that. So probably a pretty similar (laughs) way to, uh, to the way most single family homes and multifamily homes, uh, are, are built currently, just, like, with even tighter use of, of, uh, of materials. Yeah, which is per- (laughs) which perhaps says something not super nice about the way that you guys build your homes, but...

   15. DPDwarkesh Patel36:56

       (laughs) Um, uh, okay, so Ch- China is the third, uh, the third one here. There's been a lot of talk about a potential real estate bubble in China, that they're building housing in places that people don't really need it, and of course, the dem- maybe the demographics aren't there to support the, the demand. Uh, wh- what do you think of all this talk? I, I, I don't know if you're, like, uh, familiar with it but w- i- is there a real estate bubble that's created by all this competence in building?

   16. BPBrian Potter37:23

       Oh, gosh. Yeah. I have, I have, yeah, no idea. Like you, I've s- I've definitely heard talk of it and I've seen, like, you know, the little YouTube clips of, like, them knocking down all these towers that, (laughs) that it turns out they didn't need, or, like, the developer couldn't, can't finish or, or whatever. I, yeah, I don't know a, uh, a huge amount about that. In general, I, I wish I knew a lot more about how things are built in China but the information, it's, it's in general so opaque and I generally kind of assume that any sort of particular piece of data (laughs) that comes out of China has, like, giant error bars on it as to whether it's, whether it's true or not, or, like, what the context surrounding it is. So, in general, uh, I, uh, do not have a (laughs) an informed opinion about that.

   17. DPDwarkesh Patel38:09

       And then, uh, this is the second part of Simon's question. Does greater competence in, um, being able to build stuff, does that translate into other good outcomes for these countries, like higher GDP or lower rents or other, other kinds of important outcomes?

   18. BPBrian Potter38:22

       That's a good question. Japan is an interesting example where basically people point to it as a sort of example of, you know, here's a country that builds, like, huge amounts of housing and they don't have housing cost increases, and in general, we should expect that dynamic to be true, right? We should... There's no reason to not think that housing cost is essentially a supply and demand problem. If you built as much as people wanted, the, the, the, the cost would drop. Like, I have no reason to not think that's true. There is a little bit of, uh, of, of, of evidence that sort of suggests that the cost of housing will kind of ri... It's just impossible to build housing enough that, to overcome this sort of mechanic where the cost of it tend to match, to, tend to rise to, um, to whatever people's income level are. Like, the, like, the, the, the peak in, or, or the sort of flattening of, uh, uh, housing costs in Japan also parallel when people basically stopped getting, uh, raises and, and, and incomes stopped rising in, uh, in, uh, Japan. So I don't have a good sense of, um, if... again, if it ends up being just, uh, k- just, uh, im- uh, you know, more driven by some sort of other factors. Uh, but in general I expect, uh, the very basic answer of if you build a lot more houses they will, uh, the- the housing will become cheaper.

   19. DPDwarkesh Patel39:40

       Right. Uh, speaking of how the land keeps, uh, gaining value as people's income go up, wh- what is your opinion of Georgism? Like, uh, does that kind of trend make you think that housing is a special asset that needs to be more heavily taxed because it's not a... you're not inherently, like, doing something productive just by owning land th- the way you would be if you, like, built a company or something?

   20. BPBrian Potter39:57

       Yeah. I don't have any special, uh, deep knowledge of Georgism. It's one on my, one of, on my list of, of, of topics to, uh, r- read more deeply. I mean, yeah, I do think in general, yeah, you know, taxing encourages to produce less of something, for something that you can't produce less of, it's, uh, it's a good avenue for, for something to, to tax more heavily. And then yeah, obviously if you sort of had a really high land value tax in these places that have, like, a lot of single family homes in, like, dense urban areas like Seattle or San Francisco or something like that, I think, yeah, that would probably encourage the, the land to be used, um, a lot more efficiently. Um, so it does make sense, uh, yeah, makes sense to me but I don't have any, uh, a ton of special knowledge about it.

   21. DPDwarkesh Patel40:43

       All right. Ben Coon asked on Twitter, "What, uh, construction-related advice would you give to somebody building a new charter city?"

   22. BPBrian Potter40:49

       That is interesting. I mean, I think I would just, off the top of my head, I would be interested in if you could really figure out a way to build that would, like, using the method that have, like, really high upfront costs, higher than could otherwise be justified, but if you're gonna build, you know, 10,000 buildings or whatever all at once, you could really take advantage of that. One, one kind of thing that you, you see is that a lot of... in the, in the sort of post-World War II era is that we were building such, just, huge masses, massive amounts of housing and, um, a lot of times building them all in one place, right? Like the Levittown builders were building, like, thousands and thousands of houses, like, in one big development all at once. Like, in California it was the same thing, just building, like, five or 10 or 15,000 houses in one just big, massive development. And there you end up seeing something like that where they basically build, like, this, like, little factory on their construction site and then use that to, like, fabricate all these things. And then you have, like, a r- you have almost like a reverse assembly line where, where a crew would, like, go to one house and install, like, the, the walls or whatever and then go to the next house and do the same thing. Following right behind them would be the guys doing the electrical, you know, the plumbing and stuff like that. It's like this reverse assembly line system that allowed you to sort of (laughs) get these things up, like, really, really fast, like in 30 days or something like that, um, a house, just thousands and thousands of houses, um, at once. So I think you could, you know, you'd want to be able to do something similar where you could, you know, don't just do it the way that... don't just do the construction the way that normal construction, uh, is done. Um-I mean, but that's hard, right? Because a lot of the, the, the, it, the, this, the sort of centrally planned city (laughs) doesn't ever seem... Or like, the plan- the top-down planned city never seems to do particularly well, right? Like, what's the city, Brasilia or whatever? The one that they built in... It was supposed to be a planned city in like the age of like, again, with the, you know, goes back to like the unfettered technocrat who can sort of build whatever he wants, right? A lot of times what you want is something that will, sort of responds at a low level and organically to sort of the factories as they develop, and you don't want something that's planned totally top-down, that's disconnected from all the sort of cases on the ground. And a lot of, you know, what the oppositions, or Robert Moses and something like that, ended up being that in a certain form, right? Like, he's bulldozing through these cities that are, are, are these buildings and neighborhoods that he doesn't... is not paying attention to at all. And so I think, you know, just to go back to the, the question, trying to sort of plan your city from, from the top down, I don't think that... That doesn't have like a super, (laughs) super great track record, right? In general, you want your sort of city to sort of develop a little bit more organically. I guess I would think just really, uh, you know, the- the really good sort of land use rules that are really thought through well, and encourage the things that you want to encourage, and not discourage the things that you don't wanna discourage. You know, don't have equity in zoning or whatever, allow a lot of mixed use construction and stuff like that. Um, so yeah, I guess that's a, (laughs) that's a somewhat boring answer, but probably, uh, something like that.

   23. DPDwarkesh Patel43:59

       Interesting, interesting. I, I, I guess that implies that there would be like high upfront costs to building a study, because if you need to like build 10,000 homes at once to achieve these economies of scale, then you, we would need to raise like tens of billions of dollars before you could, uh, build this charter city.

   24. BPBrian Potter44:15

       Yeah, if you were trying to, yeah, lower your costs of, of construction. But again, you know that if, if you have the setup to do that, you wouldn't necessarily need to raise it, right? Like these, these other big developments were built by developers that essentially saw an opportunity, right? They didn't require, it didn't require public funding necessarily to, to... I mean, it d- it did in the form of like, uh, loans, l- you know, loan guarantees for veterans and things like that. But, um, they didn't have to... The government didn't go and buy the land.

7. 44:45 – 1:00:51

   Automation and Revolutionary Future Technologies

   1. BPBrian Potter44:45

   2. DPDwarkesh Patel44:45

      Right. Uh, okay. So the next question is from Austin Vernon. And, uh, to be honest, I don't understand the question, but I'm, uh...

   3. BPBrian Potter44:52

      (laughs) .

   4. DPDwarkesh Patel44:52

      Uh, h- h- uh, y- you two are too smart for me. But h- hopefully you'll be able to explain the question and then also answer it. Hey guys, I hope you're enjoying the conversation so far. If you are, I would really, really appreciate it if you could share the episode with other people who you think might like it. This is still a pretty small podcast, so it's basically impossible for me to exaggerate how much it helps out when one of you shares the podcast. You know, put the episode in a gr- the group chat you have with your friends, post it on Twitter, send it to somebody who you think might like it. All of those things helps out a ton. Anyways, back to the conversation. What are your power rankings for, uh, technologies that can tighten construction tolerances? And then he gives examples like AR/VR, CNC cutting, um, and synthetic wood products.

   5. BPBrian Potter45:42

      Yeah, so this is a, this is a very interesting question. So basically, because buildings are built, you know, manually on site by hand, there's just a lot of variation in what ends up being built, right? There's that, you know, things aren't... There's, there's only so acc- so accurately that a person can put something in place if they don't have any sort of aids or stuff like that. Um, so just the placement of materials tends to be, have a lot of variation in that. And then the materials themselves kind of have a lot of variation. And um, you know, the obvious example is like wood, right? Where a, one two by four is not going to be exactly the same as the two by... The adjacent two by four. It may be warped, it may have knots in it, it may be split or something like that. And then also because these materials are sitting just outside in the elements, they sort of end up getting a lot of distortion. They either absorb moisture and sort of expand and contract, or they grow and shrink because of the heat. Um, so there's, yeah, there's just a lot of variation that goes into, uh, uh, putting a, putting a building up. And to some, you know, it's, to some extent, it probably constrains what you are, uh, what you are able to build and how effectively you're able to build it. I kind of gave an example before of, um, you know, really energy efficient buildings, and they're really hard to build in, um, on site using sort of conventional methods, because just the air sealing is quite difficult to do. You have to build it in a much more precise way than like, is typically done and is really easily, uh, achieved on site. So I guess in terms of, uh, examples of things that I, that would, uh, would make that easier. Uh, I mean, he gives some, he gives some good ones like, you know, um, engineered lumber, which is like, you take lumber and you grind it up into like strands or chips or whatever, and you basically glue it back together. Uh, which does a couple of things. It basically spreads all the knots and the defects out, so they aren't concentrated. Uh, and then it lets you get like, a lot... Everything tends to be a lot more uniform, um, when it's made like that. So like, that's a, that's a very obvious one. I mean, that's already widespread use. I don't really see that, um (laughs) , making, uh, a substantial change. I guess the one exception to that would be that there's this, uh, engineered lumber product called... Or like, mass timber elements. CLT is a big one, which is like a super plywood, where plywood is made from like tiny little sheet thin strips of, um, of wood, right? And then CLT is made from like two by fours dimensional lumber, glued in like cross-laminated layers or whatever. So it's like instead of a four by eight sheet of plywood, you have like a 12 by 40 sheet of like, dimensional lumber glued together, and you end up with a lot of the properties of, you know, engineered material, where it's really dimensionally stable, it can produce, be produced very, very...... accurately. (laughs) It's actually funny is that a lot of times, the CLT is the most accurate part of the building. And so if you're building a building with it, you tend to run into problems where the rest of the building is not accurate enough for it. So even something like steel, right? If you're building, like, a steel building, the steel is not going to be quite ex- it's not going to be, like, dead on. There's going to be, like, an, uh, inch or so off in just terms of where any given component is. And so the CLT, which is built much more accurately, actually tends to, like, show all these, where these errors are that then have to be corrected. So in some sense, accuracy is like a, is a little bit of a, or precision is, is a little bit of, like, a, of a, of a tricky thing because you can't just make one part of the process more precise. All that does is you, you, you know... In some ways, that actually makes things more difficult because then it force... Instead of... If one part is really precise that... a lot of times means that you can't make adjustments to it easily. And so if you have at least one really precise thing, it usually means you have to go and make s- you know, make compensation for something else that is not built quite as precise. Um, so it actually makes construction quite, you know, of advancing, like, precision quite a bit more complicated. AR/VR, I think, is I, I'm very bullish on that, assuming, you know, with the big caveat of that assuming they can sort of just get the basic technology working. Um, the, the basic intuition there is that, you know, right now the way that, um, pieces or, of building this put together on site is that somebody is looking at a set of plans, paper plans or, like, an iPad or something, that tells them where everything needs to go. And then so they, they, they figure that out and they take a tape measure or something like that, or, like, um, you know, uh, some other, some other method and go figure out where that is and mark it on the ground and then try to sort of get it all put in there precise. So there's all this, like, setup time it takes and, and manual location that is really quite time-consuming and, and, and error-prone. Again, there's only so accurate that a guy dragging a tape, um, 40 feet across site being held by another guy. There's only so accurate that that process can be, right? And so yeah, you can ima- it's, it's very easy for me to imagine just, um, AR or stuff like that where you could just project exactly where the components of your building need to go. That would just, A, allow you a much higher level of accuracy that you couldn't easily get using, using manual methods. And then B, just reduce all that time it takes to, like, manually measure things. So I can imagine it being, like, much, much, much faster as well. So I'm quite bullish on that at a high level. At a slightly lower level, um, I'm not... It's not obvious to me if the, uh, actual

   6. DPDwarkesh Patel51:21

      (laughs)

   7. BPBrian Potter51:21

      ... technology will be able to sort of get that level where it just, like, projects it with perfect accuracy right in front of you. Um, it may be the case that, like, a person moving their head around, uh, and constantly changing and changing their point of view, you're never gonna be like, "Locate the, uh..." You're never gonna be able to, like, project these things with, like, millimeter precision or whatever. It's always gonna be, like, a little bit jumpy or it's, you're gonna end up s- with some sort of hard limit in terms of, like, how precise you can project it. I wouldn't bet on that. I just, my sense is that locator technology will get good enough, but I don't have any principled (laughs) , I don't have any principled reason for believing that. And then, and then the other thing is that being able to sort of take advantage of that technology would, um, would require you to have like a really, really accurate model of your building that, like, locates where every single element is precisely and exactly what its tolerances are. Right now, um, buildings aren't designed like that. Um, they are built using a sort of comparatively sparse, uh, set of drawings that leaves a lot to sort of be interpreted by the people on site doing the work. Um, and efforts that have tried to make these models, like, uh, really, really, really precise a lot of times st- have not really paid off. You can get, you can get returns on it if you're building something, like, really, really complex where there's a lot of, where there's a much higher premium to, uh, being able to make sure you don't make any errors or whatever (laughs) . But for, like, a simple building, like a house, the, the, the re- um, the returns to it just, just aren't there, and so you see really comparatively sparse drawings. Um, and so, you know, whether the, whether it's gonna be able to work worth this upfront cost of developing this, like, really, really complex, very precise, uh, model, um, where exactly every component is, is kind of to be determined. There are some interesting companies that are, like, trying to, I think, move in, in this direction, but they're basically making it a lot easier to, like, draw these things really, really precisely that would just be, uh, you know, have every single component of exactly where it is. Um, so I'm optimistic about that as well, but it's a little bit TBD.

   8. DPDwarkesh Patel53:32

      All right. Th- this raises a question that I actually wanted to ask you which is, in your post about why there aren't, um, auto- automatic bricklayers, um, it was a really interesting post. And somebody left in, um, a, an interesting comment which was that bricks were designed to be handled and assembled by humans. And, uh, and then I think you... L- let me just pull this up. You, you left a response to that which I thought was really interesting. You said, "The example I always reach for is with steam power and electricity where replacing a steam engine with an electric motor in your factory didn't do much for productivity. Improving factory output required totally redesigning the factory around the capability, capabilities of electric motors." And then so I was kinda curious, if you apply that analogy to construction, then what does that look like for construction? Like, what, what is, uh, uh, what is a house-building process or a building-building process that it takes, like, automation and these other kinds of tools, uh, into account? How, how would that change how buildings are built and, like, how they end up looking in the end?

   9. BPBrian Potter54:32

      I, I think that's a good question. And yeah, I think that's one big part of, you know, one big component of, like-... construction productivity, is everything is designed and is evolved over, you know, 100 years and 200 years or whatever, to be easy for a guy or, you know, person on the site to manipulate by hand. Um, so like bricks are roughly the, yeah, size and shape and weight that a person can move it easily around. Uh, dimensional lumber is the same. It's the size and shape and weight that a person can, can, can move around easily. Kind of all construction materials are like this. And the way that they attach together and stuff is the same. It's all designed so that a person on site can sort of put it all together with as, as comparatively little effort as possible. And then you... But yeah, what is, um, what is easy for a person to do is usually not what is easy for a machine or a, a robot to do, right? You typically need to sort of redesign what... you know, think about what your end goal is and then redesign the mechanism for accomplishing that in terms of like what is easy for it to get to make a machine to, to do. I mean, the obvious example here is like it's way easier to build a, like wagon or a cart that like rolls than it is to build like a mechanical set of legs that (laughs) mimics a human's movement, right? Um, that's just way, way, way easier. So yeah, I do think a big part of advancing construction productivity is to basically figure out how to redesign these building elements in a way that is really easy for a machine to produce and a machine to put together. And, and, and one, I think reason that we haven't... Yeah, haven't seen it, is that a lot of, a lot of the mechanization, mechanization that you see is people trying to mechanize exactly what a person does. So if a person moves like these two high floors, you know, into place, they're trying to make a machine that will move them before and then nail them together. And to do that, it's like really, really expensive. You need a lot of like really expensive industrial robots that can move exactly the way that a human moves, more or less. Um, yeah. And so I think, yeah, a lot of like, uh, what, what that kind of might look like is basically something that can be like really easily extruded by machine in something like a continuous process that wouldn't require a lot of like finicky mechanical movements. And, uh, yeah, something like that. Uh, a good example of, I think this technology, is a technology that's called like insulated metal panels, which is perhaps one of the cheapest and easiest ways to build, uh, an exterior wall. And what it is, it's just like a, um, it's just like a thin layer of steel. Uh, and then on top of that is a layer of insulation. And on top of that is another layer of steel. And then at the end, there's sort of... The steel is like extruded in such a way that it can, like... these inner panels can like lock together as they go. And it's basically the simplest possible method of constructing a wall that you can imagine almost. Um, but then that has like the structural system and the water barrier and air barrier and the insulation all just in this like one really simple, um, assembly. And then it go... when it go... when you put it together on site, it just locks together. (laughs) And of course, it doesn't... You know, there are a lot of limitations to this. Like, if you want to do anything complicated, like add windows or something like that, all of a sudden it starts to look quite a bit less good. But yeah, I think things like that, that I think are like really easy for, um, a machine to do can be put together with, like, a lot of like really... you know, don't need a lot of like precision measurement or stuff like that in the field. They can just kind of like snap together and actually line up fit together. I think that's kind of what it looks like.

   10. DPDwarkesh Patel58:15

       And then what would the houses or the buildings that are, um, built like using this method, what would they physically look like? I mean, maybe in like 50 to 100 years we'll look back on the houses we have today and we'll be like, "Oh, look at that artisanal creation-"

   11. BPBrian Potter58:26

       (laughs)

   12. DPDwarkesh Patel58:27

       ... "uh, made by humans." But then what, what is a machine that is, like designed for hu- uh, for robots first or for automation first? Like, in what interesting ways would it differ from today's buildings?

   13. BPBrian Potter58:37

       Yeah, that's a, that's a good question. I'm not especially bullish on like 3D building printing in general, but this is another example of like a build-... you know, building a build-... you know, a building using like an extrusion process that is relatively easy to, to mechanize, right? And sort of what's interesting there is that when you start doing that, a lot of these other, like, bottlenecks become unlocked a little bit. Like, it's very difficult to, um, build a building using like a lot of like curved exterior surfaces, like by using conventional methods, right? Like, you can do it, but it's quite expensive to do. Um, but it's re-... there's relatively straightforward for, uh, like a 3D printed building for that to do, right? Like it's... they can build that just as easily like it's a straight wall. Um, and so you see a lot of interesting curved architecture on these things, and you're kind of seeing that in a few other areas. There's company-... there's a company that can like build like this cool undulating façade or whatever that people kind of like. Um, and yeah, so it kind of unlocks a lot of options. So kind of machines, you know, they have their core constraints in some things that they can, can do, but they don't have a lot of the other constraints that you would otherwise see. So I think you'll kind of see a, a larger variety of, of sort of aesthetic things like that. Um, that said, at the end of the day, I think a lot (laughs) of the ways that like a house goes together or whatever is pretty well shaped to just the way that it... you know, a person living inside it would like to use. I think Stewart Brand makes this point in, um-

   14. DPDwarkesh Patel1:00:12

       Oh, oh, How Buildings Learned.

   15. BPBrian Potter1:00:13

       There we go. But he basically makes the point that a lot of people try to use, um, you know, like dome-shaped houses or octagon-shaped houses, which are, um, good because again, going back to surface area volume, they enclose the space using the least amount of material possible. So like in some theoretical sense are quite efficient, but it's actually quite inconvenient to live inside a building with like really curved walls, right?

   16. DPDwarkesh Patel1:00:37

       (laughs)

   17. BPBrian Potter1:00:37

       Like furniture doesn't fit up against it nicely and, and stuff like that. Like pictures are hard to hang on a really curved wall. Um, and so I think you would see less variation than maybe you might expect.

   18. DPDwarkesh Patel1:00:50

       Interesting.

8. 1:00:51 – 1:08:02

   3D Printer Pessimism & The Rising Cost of Labour

   1. DPDwarkesh Patel1:00:51

      And then why are you pessimistic about 3D printers? Or for, for construction, I mean.

   2. BPBrian Potter1:00:55

      Yeah, yeah, poor construction. (laughs) Oh gosh, so many reasons. Uh-

   3. DPDwarkesh Patel1:00:58

      (laughs)

   4. BPBrian Potter1:00:58

      Not, not pessimistic, but just, there's a lot of unanswered questions. I mean, so the big obvious one is, like, right now a 3D printer can basically print the walls of a building. That is a pretty small amount of, like, the value in a building, right? It's maybe the, um, 8%, 7 or 8%, something like that, probably not more than 10% of, like, the, the value in a building. 'Cause you're not in- you're not printing the foundation, you're not printing, like, the overhead vertical, you know, o- overhead spanning structure of the building. You're basically just printing the walls. And you're not even really pres- printing the second story walls that you have in multiple stories. I don't think they've quite figured that out yet. So it's a pretty small amount of value add to the building. And it's like, it's frankly a task that is relatively easy to do (laughs) by, like, manual labor, right? It's really pretty easy for, um, uh, a, a crew to, um, basically put up a, a s- a structure of a house. (laughs) It's, it's kind of a, a recurring theme in, like, in, like, mechanization or i- goes back to what I was talking about previously, where it takes, like, a lot of mechanization and a lot of expensive equipment to replace what basically, like, two or three guys can, can do on, in, you know, a day or something like that. It's the, the economics of it are, are pretty brutal. But anyway, so right now it produces a pretty small value of that. So I think the, the value of 3D printing is basically, uh, entirely predicated on how successful they are at figuring out how to, like, deliver more components of the building using their system. And there's companies that are trying to do this. There's, um, there's one that got funded not too long ago called Black Diamond where they have, they have this crazy system which is like a series of 3D printers that would act simultaneously, like each one building a separate house. And then as you progress you, like, switch out the print head for, like, a robot arm. 'Cause a 3D printer is basically like a robot arm with just a particular manipulator at the end, right? And so they switch out their print head for, like, a robot arm, and the robot arm goes in and in- installs, like, different other systems, like the windows or the mechanical systems or something like that. So if you can figure out how to do that reliably where your print head, uh, or your printing system is, like, installing a large fraction of the value of the building, it- it- it- it's not, it's not clear to me that it's gonna be economical, right? But that's o- that's, it obviously needs to reach that point. And it's not obvious to me that they have, like, that they have, they've gotten there yet. It's really quite hard to sort of get a robot to do a lot of these, these tasks. It'll kind of, kind of depend on how successful they are at doing that. And it- a lot of, for a lot of these players, it seems like they're actually moving away from that. Like I think in a recent... Um, Icon is the biggest, uh, construction 3D printer company in the US, uh, as far as I know. Um, and as far as I know they've moved away from, like, trying to, like, install lots of systems, uh, in their walls as they get printed. And they've kind of moved on to, like, having that installed separately. Uh, which I think has made it- their job a little bit easier, but it's again not quite... It's hard to see how the 3D printer can fulfill, (laughs) fulfill its promises if it can't do anything just beyond the, the vertical elements, which again, are really, for most construction are really quite cheap and simple to build.

   5. DPDwarkesh Patel1:04:20

      Now if- i- if you take a step back and talk about, like, overall h- how expensive construction is, how much of it can just be explained by, uh, the bamboo cost effect? That, like, labor costs are increasing because labor is more productive in oth- other industries and therefore construction is getting more expensive?

   6. BPBrian Potter1:04:37

      Yeah. I mean, I think that's, like, a huge, huge check of it. Yeah, labor has not, you know, the labor fraction of it has, has, uh, I think not changed appreciably enough. Um, I'm gonna ask for a fact 'cause I haven't actually verified that and I need to and (laughs) and I, and I'm remembering somebody that said, "Actually they used to be much different," um, who, who sent me some literature related to it. So slight asterisk on that, but in general yeah, the labor cost is just, has remained just a huge fraction of, of uh, of, of the overall cost of the building. Uh, things that were like labor is a huge cost of it, right, had like reliably seen their costs-

   7. DPDwarkesh Patel1:05:12

      Yeah.

   8. BPBrian Potter1:05:13

      ... continue to, to rise. I think there's, there's no reason to believe that that's not a, a big part of it.

   9. DPDwarkesh Patel1:05:18

      Mm-hmm. Now I know this sounds like a question with an obvious answer but in your post comparing the prices of construction in different, uh, countries, you, you ta- mentioned how the, the cost of labor and the cost of materials is not as big, um, a determiner of, like, how expensive it is to construct in different places, but what does matter is the, like, the amount of government involvement and administrative overhead. I'm curious why those thing, like government involvement and administrative overhead have such a high consequence on the cost of, uh, construction.

   10. BPBrian Potter1:05:50

       Yeah, that's a, that's a good question. I don't actually know if I have a, a unified theory of that. I mean ju- basically with any regulation thing it just, you know, any heavily regulated thing, it just, any particular task that you're doing takes longer and, uh, is less reliable than it would be if it was not done, right? You can't just do it as fast as on your own schedule, right? It's, you end up being bottlenecked by, by different processes, processes and reduces kind of the, it narrows your, your, your options. Uh, so yeah, in general I would expect that to, to kind of, uh, be the case, but actually don't know if I have, like, a unified theory of, of how that works beyond just it's a bunch of additional steps at any given part of the process, each of which kind of adds, adds cost. (laughs)

   11. DPDwarkesh Patel1:06:40

       Yeah. Now, uh, one interesting trend we have in the United States with, uh, construction is that a lot of it is done by Latino workers and especially, uh, by undocumented Latino workers. What is the effect of this on the price and the quality of construction? So if you have, like-... uh, uh, like a b- bunch of hardworking undocumented workers who are working for below market rates in the US. Will this have dampened the cost of construction over time? Or what do you think is going to happen?

   12. BPBrian Potter1:07:06

       So I suspect that's pr- that's probably one of the reasons why the US has comparatively low construction costs compared to other parts of the world. Like, well, I'll, I'll caveat that for residential construction, right, which is single family homes and multifamily apartment buildings, those are all built in the US out of light-framed wood and put together, like you said, by a lot of immigrant workers. And then, yeah, uh, be- because of that, it would not surprise me if those wages are a lot lower than, you know, the equivalent wage for a carpenter in Germany or something like that, and that results in our, our, the cost of our resident construction, which is very, very low by, by world standards. I, I suspect that is, um, that is a big... That's a factor in why our costs of, of resident construction a- are quite low.

   13. DPDwarkesh Patel1:08:00

       Mm-hmm. Um, now

9. 1:08:02 – 1:17:53

   AI’s Impact on Construction Productivity

   1. DPDwarkesh Patel1:08:02

      overall, it seems from your blog post that you're kind of pessimistic, or y- you don't think that, um, like different improvements in industrialization have transferred over to construction yet, but what do you think is the prospect of future advances in AI having a big impact on construction? So like, oh, you know, with computer vision and with advances in robotics, do you think we'll finally see some carryover into, um, construction productivity or, uh, or is it gonna be more of the same?

   2. BPBrian Potter1:08:30

      Yeah, it would... I think there's definitely going to be progress on, on that axis. Yeah, if you can wire up your computer vision systems to your robotic systems and your AI in such a way that your capabilities for a robot system are more expanded, uh, yeah, I kind of foresee the, you know, robotics being able to take a larger and larger fraction of the tasks done on a typical construction site. I kind of see it being kind of done in, in, um, narrow avenues that gradually expand outward. You're starting to see a lot of companies that, um, they have some robotic system that can do one particular task, but it can actually do that task, um, quite well. There's a couple different robot companies that have these little robots for drawing wall layouts on concrete slabs or whatever so you know exactly where to build your walls, which is... You would think it would not be a difficult problem in construction, but it turns out that a lot of times people put the walls in the wrong spot and then you have to go back and move them later or just basically deal with it. And so yeah, they have... It's basically a little Roomba-type device that just draws the, the um, the wall layout onto the concrete slab and all the other systems as well, like where the lines need to run through the slab and things like that. So yeah, I kind of suspect that you're just going to start to see robotics, um, and systems like that just take a larger and larger share of, uh, of the tasks on the construction site o- over time.

   3. DPDwarkesh Patel1:10:08

      Mm-hmm. Uh, what do you think of Flow? That's Adam Neumann's newest startup that is backed with $350 million from Andreessen Horowitz.

   4. BPBrian Potter1:10:17

      (laughs) Um, I do not have any, uh, strong opinions about that other than... Wow, they're really giving him another three hours, right?

   5. DPDwarkesh Patel1:10:27

      (laughs)

   6. BPBrian Potter1:10:27

      Um, yeah, I do not have any particular strong opinions about that. I will say that, um... Was it A16Z where they-

   7. DPDwarkesh Patel1:10:35

      Yeah.

   8. BPBrian Potter1:10:35

      ... were the ones that wrote the check? They made a lot... They make a lot of investments that don't make sense to me, but I'm not a venture capitalist, so there's no reason to... You know, there's no reason that my judgment would be any good in this situation. So I'm just presuming they know something I do not.

   9. DPDwarkesh Patel1:10:50

      Yeah, I'm going to be interviewing Andreessen later this month and (laughs) -

   10. BPBrian Potter1:10:54

       Oh, nice.

   11. DPDwarkesh Patel1:10:54

       ...I'm hoping I can ask him about that.

   12. BPBrian Potter1:10:56

       I would love to hear what that thing... I mean, you know, it's just... It may be just like... It may be as simple as, like, you know, VC is all about really high variance bets. There's nobody higher variance than Andreesen. (laughs)

   13. DPDwarkesh Patel1:11:07

       Yeah, yeah.

   14. BPBrian Potter1:11:09

       Then, uh, yeah. So Adam Newman. So, uh, yeah, maybe just on those terms, it makes sense.

   15. DPDwarkesh Patel1:11:16

       You had an interesting post about how a bunch of... A lot of the knowledge in the construction industry is informal and contained within Best Practices or between relationships and expectations that are not articulated all the time. It seems to me that this is also true of software in many cases, but software seems much more legible and open source, if that makes sense, than these other physical disciplines like construction, despite again, having a lot of the knowledge contained within people's minds and within the culture rather than explicitly codified somewhere. So, um, w- why do you think that construction seems more closed source than software?

   16. BPBrian Potter1:12:05

       It's interesting. I mean, to go back slightly to our products versus projects industry, right? Um, a slightly different way of thinking about that is just like craft-based industries via... Versus, again, an industrial process where a craft-based industry, basically you have... And this isn't like a dichotomy, right? This is a spectrum. But in general, like craft-based stuff, there's an expertise and judgment aspect of it that is sort of pretty well embedded in the process that you can't really well remove it. Any sort of decision at any given point requires an expert or an artisan or somebody who understands the relevant context and knows how to proceed based on the specific variables in this specific situation, where industrial process is more like, this has been figured out and this is how it works every single time.And just construction is just very, very much on the craft end of the production spectrum, where a lot, you know, the decision of, like, how to put these things together and how to wire, you know, this building or whatever, it's all left up to the sort of the expertise and the sort of judgments of the people doing sort of the installation. And what that gets you is that, again, you can ... it lets you put things together without having to do a very large amount of specifying ex- exactly what you need, right? Like the- the drawings to specify a house going together, or even something quite complicated, um, are way fewer than the drawings needed to produce a Toyota Corolla. (laughs) Um, I'm- I'm sure, for instance. Uh, you know, so, uh ... and the- and the cost required, like, you know, uh, to- to do it, like the design cost required to do it, is un- ... in terms, in proportion to, uh, to how expensive the thing is, is also- is also much lower as well. Software is perhaps, and again, I'm not an expert on software, um, development, but it's somewhat more legible by response in that the end product is like ... it's b- you know, it's very clear. You can clearly see every single part of it and how every single part of it, um, touches every single other part. Perhaps I'm- I'm overstating that. You know, I'm sure a software could say, "Well, it's actually really not super obvious how these things work and why they're done this way," or- or- or whatever. But you can see ... you can clearly s- inspect every single part of it and s- and see exactly how it does, uh, w- you know, what it does and how it connects the other- the oth- the other part of it. You can't really ... I mean, you can't really do that with a building. I guess I would also maybe say that there's ... this is maybe an interesting question. With- with- with a building, it's not necessarily obvious how it got to the point that it did when it was- when it was put together. Um, and a lot of times with, like, physical things, it ends up, even if you have the object, it is unclear how- what the process was to create it. And so a lot of times what you see is that even with, like ... this comes up with, like, industrial espionage, right? Where somebody who's trying to, like, steal some particular thing or whatever, a lot of times that doesn't help them as much as they would think to try to, like, recreate it. A lot of times they have to basically go through the entire process of figuring out how to make it, and it takes them just as long to do it as it did the original people doing the development. You saw this with, like, the development of the atomic bomb, for instance, where, like, the people who, like, stole the- (laughs) stole the plans for how to make it or who had information on, like, exactly how their system, the- the bomb worked or whatever, basically took as long to figure out how to make it as the US did. So just a physical object, just the process being used to make it is not necessarily just super, super legible. It tends to be a little bit hidden. I ... and I was going to say that I for ... perhaps that is not as true for- for software, but I actually real- ... I don't actually know and it's- it's very plausible to me that you could have some piece of software that was written. I- I guess I would maybe ask you this question. Is it possible that you'd have some piece of software that was written and then just it'd be utterly inscrutable as to how it- it came together, uh, and how you could like, you know, duplicate a similar th- you know, a s- a similar piece of software? Or is that like a category error?

Episode duration: 2:25:57