Dwarkesh Podcast
00:06:03 · Robert Caro
[00:06:03] S1: They also say, Robert, because of all these safeguards, Robert Moses could not operate in the same way today. But the fact is when Robert Moses started, there were also safeguards. People were also saying, you can't build anything. This was in the 1920s. The people were saying government can't build anything. There are too many blocks. Nothing's getting built. We need highways. It's the age of cars. Everyone's getting automobiles, but we have the same old roads and nobody can build them because the property owners don't want the roads built through their land. Robert Moses comes along and all of a sudden those things vanish because he... Despite all the safeguards and everything else, he just builds. So doing the power broker, it was just fascinating to me to see how he gets around everything. But it doesn't matter what it is. He's gonna build a beach at Jones Beach. He sees it one day. He says, this is the greatest beach I've ever seen. It's only 27 miles from New York City. If I can build a road out here, I will give the people of New York City this great beach. But it seems impossible to build that road because the property owners don't want the road going through their farms and property. But he basically says, I'm gonna build it. And to watch him do it, to overcome each obstacle, there's so many. And then all of a sudden it's open and you have Jones Beach and you have it accessible to New York. It's an act of political genius, and it's also an act of will that I'm gonna do it. That's fascinating to learn about.
00:08:07 · Dwarkesh Patel, Robert Caro
[00:08:07] S0: And so you think today, if there was a Mosesite figure, they would figure out a way to work around everything that's been set up? [00:08:14] S1: Well, the answer to that is when Moses was building, they were saying the same thing. Nobody's ever gonna build, we need roads for the automobiles, but they'd have to go through hundreds, dozens of farms, scores of property, no one's ever gonna be able to do it, but he comes along and doesn't. So today someone would come along and do it. [00:08:44] S0: This I think is a profound question for democracies which is how do we end up in a place which is not the extremes of 50s or 60s Moses but not also where we are now. So where I live in California, we've spent, I don't know, $15,000,000,000 over the last decade and we haven't gotten a single new line or single new mile of railway that those $15,000,000,000 have been spent to bill. Whereas a different system, you could say China today is like New York City in the forties and fifties, where they have a very Moses type attitude and they put up tens of thousands of miles of railways over the last ten, twenty years. But of course at the cost of similar amounts of displacements of people and recklessness. So how do we land in a good spot between either extreme?
00:09:42 · Robert Caro
[00:09:42] S1: I'm not sure there is a good spot. What you just said is the situation. To build anything on a large scale, you're confronted with dozens or scores or perhaps hundreds of property owners who don't want it built there. So the question is how does anything get built? Then someone comes along and can do it. Now, when Moses, when Robert Moses started, it seemed impossible, impossible to build roads out into Long Island because the road would have to cross scores of properties. Most of which the people didn't want the roads coming and had the power to stop them then because he didn't have the power of eminent domain then. But Moses said, I'm going to build it. And he would find a way to overcome each difficulty. Now you say, what am I trying to say by that? Trying to say that every once in a while, comes along who sort of a a genius, if that's the right word, at getting big things built for the public. Decades may pass when there's no one like that. And you say everything's stalled, nothing nothing's gonna be... Nothing big is ever gonna be built again. And then someone suddenly comes along and all of a sudden, things get built. That's the nature of political genius. I happened to be writing about two guys, Lyndon Johnson and Robert Moses, who in both cases say Johnson with civil rights and Moses with, public works, where people were saying you can't do this, they can't... But they but they did it.
00:11:45 · Dwarkesh Patel, Robert Caro
[00:11:45] S0: Correct me if I'm putting words in your mouth, but if I may attempt an interpretation of what you said. Yes. It sounds like you think we need a little Moses today. [00:12:00] S1: Oh, a great insight. I sometimes wonder about just what you said. And the answer is, well, I certainly don't want someone who's ruthless and disregarding public opinion as a Robert Moses. But at the same time, you need someone who's gonna find a way to get things built. And he's gonna have to find a way despite the fact that so many people don't want it built. [00:12:42] S0: A lot of people who haven't read the book have this impression that you think everything Moses did was bad and that we need to make sure that we can never have a Moses again. Yes. And so it's interesting to hear you say that we need a little Moses again.
00:12:57 · Robert Caro
[00:12:57] S1: Well, when people say to me everything was bad, then he did. What I say is you didn't read the book because certainly the first parts of the book, well, hundreds of pages are about a young idealist who sees this great need for recreation, for beaches on Long Island, for the city people, and is going to get away to make them accessible to the... To people in the city. You say, and it's an act of genius and willpower the way he does it. And that's great. Now, does it become, does it evolve into where Moses listens to nobody and the only thing that matters is what he wants to get built? It does evolve into that. He does change. But the young Moses, the idealist, who before he's building is an expert in civil service, technicalities of civil service, and devotes all his energies to trying to reform the civil service and make it, that people get promoted on merit instead of political pull. And he worked so hard in that. You see a real idealism. And you see an idealism in his early works, you know, when you say, I wanna make these beaches accessible to New York. Now does that evolve into where I think there ought to be a highway here no matter what anybody says, and no matter how many thousands of people are in the way, like the crossroads of choice by Ali Victim? It does. He changes. That to me is the story of the... A big part of the story of the power broker, how this guy who's in the beginning, a young idealistic guy who wants to build, to make lives better for people, to give them access to the ocean beaches and all, changes into someone who just wants to build because he thinks something should be built no matter whatever anyone else says, and is gonna do it no matter how many thousands, hundreds, thousands people have to evict, no matter how bad an idea it is, like the Cross Bronx Expressway. People are telling him not to do that before he does it. The minute this road opens, it's the world's longest traffic jam. Not thousands, but tens of thousands of people were evicted, thrown out of their homes by him without any relocation provisions. So you say, in a way, it's a story of how do you give somebody power without the power in a way hardened, I don't want to use the word corrupting him, but hardening him to anyone who opposes him, to any opposition.
00:16:27 · Dwarkesh Patel, Robert Caro
[00:16:27] S0: Coding agents work best when they get fast, detailed feedback. My sponsor, Antithesis, gives your agents some of the richest feedback there is by running every change they make through thousands of hostile simulations. This usually isn't feasible. Simulation testing is hard. It normally requires rebuilding your software around a fake networking clock among other things. So most teams don't bother. Even if they do, getting enough coverage typically takes hours. So they run these simulations overnight, by which point all the day's changes are jumbled together and it's harder for the agents to tell which PR broke something. Antithesis solves both problems. It runs your software inside a fully deterministic simulated computer so you don't have to rebuild anything yourself. And it explores fast enough that you can test every PR. On a recent project, it packed fifteen hours of simulation testing into each fifteen minute run. That speed comes from both parallelism and branching. Because it's deterministic, whenever Antithesis finds an interesting state, it can save that exact moment and branch off hundreds of different futures. It never wastes time replaying events, hoping to stumble across that state again. And it doesn't just explore randomly. Antithesis systematically injects faults and tweaks your system in the ways most likely to break. So every minute in Antithesis is one of the unluckiest hours your software will ever experience. Learn more at antithesis.com/dwarkesh. [00:17:54] S1: You're [00:17:58] S0: writing the book. You start writing it in '66, '67?