1
00:00:00,080 --> 00:00:02,960
Speaker 1: If you love iPhone, you'll love Apple Card. It's the

2
00:00:03,040 --> 00:00:06,920
Speaker 1: credit card designed for iPhone. It gives you unlimited daily

3
00:00:07,040 --> 00:00:10,000
Speaker 1: cash back that can earn four point four zero percent

4
00:00:10,160 --> 00:00:12,960
Speaker 1: annual percentage yield. When you open a high Yield savings

5
00:00:13,000 --> 00:00:16,280
Speaker 1: account through Apple Card, apply for Applecard in the wallet

6
00:00:16,280 --> 00:00:19,520
Speaker 1: app subject to credit approval. Savings is available to Apple

7
00:00:19,560 --> 00:00:22,800
Speaker 1: Card owners subject to eligibility. Apple Card and Savings by

8
00:00:22,840 --> 00:00:26,120
Speaker 1: Goldman Sachs Bank USA, Salt Lake City Branch Member FDIC

9
00:00:26,520 --> 00:00:29,280
Speaker 1: terms and more at applecard dot com. When you pop

10
00:00:29,320 --> 00:00:31,240
Speaker 1: a piece of cheese into your mouth, you're probably not

11
00:00:31,360 --> 00:00:34,280
Speaker 1: thinking about the environmental impact. But the people in the

12
00:00:34,360 --> 00:00:37,680
Speaker 1: dairy industry are. That's why they're working hard every day

13
00:00:37,720 --> 00:00:40,760
Speaker 1: to find new ways to reduce waste, conserve natural resources,

14
00:00:40,800 --> 00:00:44,360
Speaker 1: and drive down greenhouse gas emissions. How is US Dairy

15
00:00:44,400 --> 00:00:48,520
Speaker 1: tackling greenhouse gases? Many farms use anaerobic digesters to turn

16
00:00:48,560 --> 00:00:53,080
Speaker 1: the methane from manure into renewable energy that can power farms, towns,

17
00:00:53,120 --> 00:00:57,240
Speaker 1: and electric cars. Visit us Dairy dot COM's Last Sustainability

18
00:00:57,280 --> 00:00:58,000
Speaker 1: to learn more.

19
00:00:58,960 --> 00:01:02,600
Speaker 2: Here's a little secret. Most smartphone deals aren't that exciting.

20
00:01:02,920 --> 00:01:06,560
Speaker 2: To be honest, they're barely worth mentioning. But then there's

21
00:01:06,600 --> 00:01:09,920
Speaker 2: AT and T and their best deals. Those are quite exciting.

22
00:01:10,319 --> 00:01:13,039
Speaker 2: They're the kind of deals that are really worth talking about,

23
00:01:13,560 --> 00:01:16,360
Speaker 2: like their deal in the new Samsung Galaxy Z flip six.

24
00:01:16,760 --> 00:01:19,600
Speaker 2: With this deal, you can trade in your eligible smartphone,

25
00:01:19,680 --> 00:01:22,960
Speaker 2: any year, any condition for a new Samsung Galaxy Z

26
00:01:23,080 --> 00:01:23,680
Speaker 2: flip six.

27
00:01:24,120 --> 00:01:26,160
Speaker 3: It's so good, in fact, it will have.

28
00:01:26,280 --> 00:01:30,640
Speaker 2: You shouting from the rooftops. So get yourself down a

29
00:01:30,640 --> 00:01:33,360
Speaker 2: street level and learn how to snag the new Samsung

30
00:01:33,400 --> 00:01:36,880
Speaker 2: Galaxy Z flip six on AT and T and maybe

31
00:01:36,880 --> 00:01:39,320
Speaker 2: grab a ladder on the way home. AT and T

32
00:01:39,720 --> 00:01:44,119
Speaker 2: connecting changes everything requires trade in a Galaxy s Note

33
00:01:44,200 --> 00:01:46,960
Speaker 2: or Z series smartphone. Limited time offer two hundred and

34
00:01:46,959 --> 00:01:49,880
Speaker 2: fifty six gigabytes for zero dollars. Additional fees, terms and

35
00:01:49,920 --> 00:01:53,360
Speaker 2: restrictions apply. See att dot com, slash Samsung or visit

36
00:01:53,400 --> 00:01:54,920
Speaker 2: an AT and T store for details.

37
00:01:57,840 --> 00:02:00,400
Speaker 4: As a United Explorer Card member, you can earn fifty

38
00:02:00,440 --> 00:02:04,320
Speaker 4: thousand bonus miles plus look forward to extraordinary travel rewards,

39
00:02:04,440 --> 00:02:06,960
Speaker 4: including a free checked bag, two times the miles on

40
00:02:07,040 --> 00:02:09,680
Speaker 4: United purchases and two times the miles on dining and

41
00:02:09,760 --> 00:02:13,079
Speaker 4: at hotels. Become an explorer and seek out unforgettable places

42
00:02:13,080 --> 00:02:16,520
Speaker 4: while enjoying rewards everywhere you travel. Cards issued by JP

43
00:02:16,600 --> 00:02:20,600
Speaker 4: Morgan Chase Bank NA Member FDIC subject to credit approval offer,

44
00:02:20,680 --> 00:02:22,480
Speaker 4: subject to change terms apply.

45
00:02:31,360 --> 00:02:34,160
Speaker 1: Hey, am I speaking to the real Jorge today?

46
00:02:34,240 --> 00:02:35,120
Speaker 3: Who else could it be?

47
00:02:35,320 --> 00:02:37,600
Speaker 1: I don't know. It could be the simulated Jorge or

48
00:02:37,639 --> 00:02:43,919
Speaker 1: an AI generated Horhea Jorge GPT Yes, chat Horge. Would

49
00:02:43,960 --> 00:02:46,320
Speaker 1: it make a difference. That sounds like something the simulated

50
00:02:46,400 --> 00:02:47,160
Speaker 1: Jorge would say.

51
00:02:47,880 --> 00:02:50,480
Speaker 3: I kind of wish I had simulated Horhes. Then I

52
00:02:50,680 --> 00:02:53,200
Speaker 3: might avoid a lot of mistakes I make. Or they

53
00:02:53,200 --> 00:02:54,760
Speaker 3: can do all the work while I sleep in.

54
00:02:54,840 --> 00:02:57,040
Speaker 1: Why do you think simulated Horges are less likely to

55
00:02:57,040 --> 00:02:57,680
Speaker 1: make mistakes?

56
00:02:57,800 --> 00:02:57,920
Speaker 4: No?

57
00:02:58,240 --> 00:02:59,679
Speaker 3: I mean they would do the mistakes and then I

58
00:02:59,720 --> 00:03:01,880
Speaker 3: would learn from there. That's the idea.

59
00:03:01,960 --> 00:03:02,160
Speaker 5: Right.

60
00:03:02,320 --> 00:03:03,799
Speaker 1: That does sound useful, But I think you have to

61
00:03:03,880 --> 00:03:06,800
Speaker 1: be careful about the sim Jorges organizing and rising up

62
00:03:06,800 --> 00:03:07,280
Speaker 1: against you.

63
00:03:07,480 --> 00:03:09,840
Speaker 3: Oh do you think they would form their own union

64
00:03:10,720 --> 00:03:11,680
Speaker 3: or a revolt? Do you mean?

65
00:03:11,840 --> 00:03:14,640
Speaker 1: Yeah, either mutiny or fair wages? Either one.

66
00:03:14,840 --> 00:03:17,279
Speaker 3: Well, I could just pay them in simulated money, I guess.

67
00:03:17,080 --> 00:03:18,600
Speaker 1: As long as they can use that to feed their

68
00:03:18,639 --> 00:03:20,480
Speaker 1: simulated children. I bet they'd be happy.

69
00:03:20,560 --> 00:03:23,919
Speaker 3: Oh no, I definitely provide simulated benefits too. The whole

70
00:03:23,960 --> 00:03:25,720
Speaker 3: simulated family gets a bonus.

71
00:03:25,800 --> 00:03:38,160
Speaker 1: You're not a good employer, but you can simulate one.

72
00:03:43,440 --> 00:03:43,600
Speaker 5: Hi.

73
00:03:43,640 --> 00:03:46,280
Speaker 3: I'm Jorge mc cartoonists and the author of Oliver's Great

74
00:03:46,280 --> 00:03:46,960
Speaker 3: Big Universe.

75
00:03:47,160 --> 00:03:50,280
Speaker 1: Hi. I'm Daniel. I'm a particle physicist and a professor

76
00:03:50,280 --> 00:03:55,000
Speaker 1: at UC Irvine, and I am constantly simulating crazy conditions.

77
00:03:55,200 --> 00:03:57,240
Speaker 3: You mean in your life or in your work.

78
00:03:57,960 --> 00:03:59,880
Speaker 1: Well, work is a big part of my life. But yeah,

79
00:04:00,080 --> 00:04:04,520
Speaker 1: my job involves simulating collisions at very high energies all

80
00:04:04,560 --> 00:04:04,960
Speaker 1: the time.

81
00:04:05,640 --> 00:04:08,040
Speaker 3: And you also actually do them right, You actually collide

82
00:04:08,040 --> 00:04:09,600
Speaker 3: things at the large Hadron collider.

83
00:04:09,760 --> 00:04:12,680
Speaker 1: That's right. We both collide particles together in real life,

84
00:04:12,840 --> 00:04:16,000
Speaker 1: and we simulate what would happen if we collided particles

85
00:04:16,080 --> 00:04:19,719
Speaker 1: under various different potential laws of the universe to see

86
00:04:19,760 --> 00:04:22,839
Speaker 1: what might happen. Will the Earth get gobbled up? Will

87
00:04:22,839 --> 00:04:25,200
Speaker 1: it create a black hole that destroys the Earth or not?

88
00:04:25,640 --> 00:04:26,400
Speaker 1: Let's find out.

89
00:04:26,600 --> 00:04:29,800
Speaker 3: Does that mean your whole career is a simulation or

90
00:04:29,839 --> 00:04:30,640
Speaker 3: your whole life?

91
00:04:30,839 --> 00:04:33,320
Speaker 1: You know, our computers are not fast enough to keep

92
00:04:33,400 --> 00:04:36,320
Speaker 1: up with reality. So while we generate lots and lots

93
00:04:36,360 --> 00:04:40,040
Speaker 1: of simulated collisions. The real collider has generated more collisions

94
00:04:40,080 --> 00:04:41,440
Speaker 1: than we could ever simulate.

95
00:04:42,160 --> 00:04:43,919
Speaker 3: But how do you know, Daniel, that we're not in

96
00:04:44,000 --> 00:04:46,960
Speaker 3: a simulation right now? Like you might think you're doing experiments,

97
00:04:47,000 --> 00:04:49,520
Speaker 3: but really you're just inside of a video game somewhere.

98
00:04:49,600 --> 00:04:51,240
Speaker 1: Well, I want to find the cheap goods, then, well I.

99
00:04:51,240 --> 00:04:52,719
Speaker 3: Think if you had found them by now, you probably

100
00:04:52,839 --> 00:04:54,080
Speaker 3: would have a noble price.

101
00:04:54,200 --> 00:04:57,200
Speaker 1: Right, I'm hoping smashing particles together gives me the cheap.

102
00:04:57,040 --> 00:04:59,840
Speaker 3: Goods, and then that opens up the real boss level.

103
00:05:00,920 --> 00:05:02,880
Speaker 1: Where I fight the simulated army of Joges.

104
00:05:03,000 --> 00:05:07,360
Speaker 3: No, he fight the real joege Bo. That's the real boss.

105
00:05:08,040 --> 00:05:10,560
Speaker 3: But anyways, welcome to our podcast, Daniel and Jorge Explain

106
00:05:10,640 --> 00:05:13,400
Speaker 3: the Universe, a production of iHeartRadio.

107
00:05:12,839 --> 00:05:15,159
Speaker 1: In which we use our tiny little minds to try

108
00:05:15,160 --> 00:05:18,400
Speaker 1: to understand the vast universe. We hope to build in

109
00:05:18,520 --> 00:05:22,680
Speaker 1: your head a simulation of sorts, one that describes the

110
00:05:22,680 --> 00:05:25,719
Speaker 1: way the real universe works out there. We hope to

111
00:05:25,960 --> 00:05:28,880
Speaker 1: encode into your brain some laws of physics that will

112
00:05:28,880 --> 00:05:31,919
Speaker 1: help you understand how the real universe out there is

113
00:05:32,120 --> 00:05:35,560
Speaker 1: smashing and bashing to create our Bonker's reality.

114
00:05:35,760 --> 00:05:38,240
Speaker 3: That's right. The universe is doing all kinds of amazing

115
00:05:38,320 --> 00:05:41,280
Speaker 3: and awe sometimes and saying things out there in reality.

116
00:05:41,600 --> 00:05:44,960
Speaker 3: And so it's our job as humans and as scientists

117
00:05:45,000 --> 00:05:47,800
Speaker 3: to understand what's going on and to ask questions, to

118
00:05:47,880 --> 00:05:50,320
Speaker 3: probe into the true answers to why things are the

119
00:05:50,320 --> 00:05:50,680
Speaker 3: way they are.

120
00:05:50,880 --> 00:05:53,320
Speaker 1: And the classical way that science does this is with

121
00:05:53,560 --> 00:05:57,960
Speaker 1: theories and experiments, hypotheses and tests. You have an idea,

122
00:05:58,040 --> 00:06:00,240
Speaker 1: you go and see out there in the universe, if

123
00:06:00,240 --> 00:06:03,000
Speaker 1: it works, you predict something happens, and you go and

124
00:06:03,080 --> 00:06:05,880
Speaker 1: check to see if it does. But the modern scientific

125
00:06:05,920 --> 00:06:08,640
Speaker 1: method has a third way, which lives sort of uncomfortably

126
00:06:08,680 --> 00:06:11,160
Speaker 1: between theory and experiment.

127
00:06:11,360 --> 00:06:14,039
Speaker 3: Oh why is it uncomfortable? Is it like uncomfortable awkward

128
00:06:14,120 --> 00:06:16,040
Speaker 3: or uncomfortable, like physically uncomfortable.

129
00:06:16,080 --> 00:06:17,880
Speaker 1: It's a little bit uncomfortable for those of us who

130
00:06:17,880 --> 00:06:20,120
Speaker 1: specialize in not to know where we fit into the

131
00:06:20,160 --> 00:06:23,920
Speaker 1: picture of science. Some people consider me an experimental physicist.

132
00:06:24,080 --> 00:06:26,279
Speaker 1: Some people are like, nah, he mostly runs simulation, so

133
00:06:26,320 --> 00:06:28,719
Speaker 1: he's really a theorist. So you can be sort of

134
00:06:28,760 --> 00:06:32,080
Speaker 1: like uncomfortably between two different communities. If you do a

135
00:06:32,080 --> 00:06:33,760
Speaker 1: lot of simulations.

136
00:06:33,720 --> 00:06:36,440
Speaker 3: You start your own simulated community. Can you be like

137
00:06:36,480 --> 00:06:38,920
Speaker 3: a simulating physicist.

138
00:06:38,600 --> 00:06:42,560
Speaker 1: A simulator or a simulate trist. That sounds not safe

139
00:06:42,560 --> 00:06:42,920
Speaker 1: for work.

140
00:06:42,920 --> 00:06:47,279
Speaker 3: Actually, yeah, I don't know what you mean, but I'll

141
00:06:47,320 --> 00:06:48,080
Speaker 3: take your word for it.

142
00:06:48,200 --> 00:06:51,360
Speaker 1: You know, academic communities change pretty slowly, and for example,

143
00:06:51,520 --> 00:06:54,960
Speaker 1: in departments of physics, people tend to hire people that

144
00:06:55,000 --> 00:06:57,839
Speaker 1: are like them. The experimentalists get to hire somebody. They

145
00:06:57,880 --> 00:07:01,719
Speaker 1: want to hire somebody who's a blue blooded experimentalists down

146
00:07:01,760 --> 00:07:04,240
Speaker 1: to the core. So if you work at the intersection

147
00:07:04,400 --> 00:07:07,080
Speaker 1: of fields, you do some experiments, you do some theory,

148
00:07:07,120 --> 00:07:09,800
Speaker 1: maybe even do some computer science and machine learning, then

149
00:07:09,840 --> 00:07:11,560
Speaker 1: you don't necessarily have a home, you don't have a

150
00:07:11,640 --> 00:07:13,240
Speaker 1: tribe that's going to go to bat for you to

151
00:07:13,280 --> 00:07:15,640
Speaker 1: get hired. So it's sort of about the sociology of

152
00:07:15,680 --> 00:07:17,320
Speaker 1: science as a real practice.

153
00:07:17,720 --> 00:07:19,360
Speaker 3: Well I think that kind of makes sense, right, Like

154
00:07:19,440 --> 00:07:23,080
Speaker 3: why hire a simulating physicist when you can just simulate one?

155
00:07:23,640 --> 00:07:24,800
Speaker 3: Why go to all the trouble?

156
00:07:24,920 --> 00:07:27,320
Speaker 1: You know, I think that's true. If we could simulate physicists,

157
00:07:27,360 --> 00:07:29,240
Speaker 1: we could get a lot more done. But we're not

158
00:07:29,360 --> 00:07:32,520
Speaker 1: quite there yet. Human physicists still have a little bit

159
00:07:32,520 --> 00:07:33,040
Speaker 1: of an edge.

160
00:07:33,240 --> 00:07:37,400
Speaker 3: Yeah, maybe till next week when chat GPT catches up

161
00:07:37,440 --> 00:07:38,760
Speaker 3: and starts doing physics.

162
00:07:39,000 --> 00:07:41,440
Speaker 1: I mean, have you ever asked chat GPT a physics question.

163
00:07:41,760 --> 00:07:43,880
Speaker 1: You don't get physics out, that's for sure.

164
00:07:44,040 --> 00:07:46,480
Speaker 3: But anyways, it is an interesting universe because I guess

165
00:07:46,520 --> 00:07:49,320
Speaker 3: sometimes there are questions. You can't just go out there

166
00:07:49,360 --> 00:07:51,400
Speaker 3: and try for yourself in the universe, right.

167
00:07:51,280 --> 00:07:54,240
Speaker 1: That's right. Simulation has emerged in the past fifty years

168
00:07:54,240 --> 00:07:57,080
Speaker 1: as an extraordinarily powerful tool as a little bit of

169
00:07:57,080 --> 00:07:59,360
Speaker 1: experiment and a little bit of theory, and it lets

170
00:07:59,400 --> 00:08:02,960
Speaker 1: us answer questions that we otherwise could not answer. It

171
00:08:03,040 --> 00:08:06,240
Speaker 1: really is a completely new tool in the science tool.

172
00:08:06,040 --> 00:08:08,480
Speaker 3: Belt, although I would argue it's maybe one of the

173
00:08:08,520 --> 00:08:11,400
Speaker 3: oldest tools in science. And so today on the podcast,

174
00:08:11,400 --> 00:08:20,600
Speaker 3: we'll be asking the question why do scientists do simulations?

175
00:08:20,880 --> 00:08:22,640
Speaker 3: Why do scientists do anything.

176
00:08:23,240 --> 00:08:25,880
Speaker 1: Other than the obvious that simulations are so much fun.

177
00:08:26,120 --> 00:08:28,360
Speaker 1: You get to build your own little universe. You are

178
00:08:28,440 --> 00:08:31,080
Speaker 1: the creator and god of that simulated universe.

179
00:08:31,320 --> 00:08:35,280
Speaker 3: Oh boy, is that that the ultimate goal? There? To

180
00:08:35,360 --> 00:08:40,000
Speaker 3: be gods? No, you know, you don't have to get

181
00:08:40,000 --> 00:08:41,920
Speaker 3: a degree for that. You could just buy some legos.

182
00:08:43,600 --> 00:08:43,760
Speaker 6: Oh.

183
00:08:43,800 --> 00:08:45,560
Speaker 1: I've been doing that since I was a little kid,

184
00:08:45,600 --> 00:08:48,240
Speaker 1: I just want more and more powerful simulations.

185
00:08:48,800 --> 00:08:51,640
Speaker 3: You want more powerful legos, smaller legos.

186
00:08:51,760 --> 00:08:53,959
Speaker 1: Jokes aside. There is a real sense of power when

187
00:08:54,000 --> 00:08:57,240
Speaker 1: you create a simulated universe because you are deciding what

188
00:08:57,320 --> 00:08:59,839
Speaker 1: the laws of physics are in that universe, what part

189
00:09:00,280 --> 00:09:02,480
Speaker 1: do they have, how do they interact, and then you

190
00:09:02,520 --> 00:09:04,640
Speaker 1: get to see how it all plays out.

191
00:09:04,880 --> 00:09:07,240
Speaker 3: M Yeah, I sort of get that. I mean I

192
00:09:07,280 --> 00:09:09,640
Speaker 3: write a lot, I create characters, and I sort of

193
00:09:09,640 --> 00:09:12,079
Speaker 3: build my own world. What's the difference.

194
00:09:12,240 --> 00:09:15,760
Speaker 1: Yeah, you could think of fiction as simulated human interaction

195
00:09:15,960 --> 00:09:18,640
Speaker 1: and lives. Right, we're exploring what it would be like

196
00:09:18,720 --> 00:09:20,920
Speaker 1: to be in those situations.

197
00:09:21,200 --> 00:09:23,199
Speaker 3: Yeah. Wait, did you just say your work is fiction?

198
00:09:23,559 --> 00:09:27,760
Speaker 1: Simulations are definitely fiction. Sometimes they align with reality, and

199
00:09:27,800 --> 00:09:30,720
Speaker 1: one deep question is how well they align. What lessons

200
00:09:30,760 --> 00:09:34,440
Speaker 1: you can learn from your simulated fiction that carry over

201
00:09:34,679 --> 00:09:35,960
Speaker 1: into the real world.

202
00:09:36,360 --> 00:09:40,679
Speaker 3: Interesting? So your research is science fiction, is what you're saying.

203
00:09:43,520 --> 00:09:45,959
Speaker 1: You know, I've always argued that there's a strong connection

204
00:09:46,040 --> 00:09:49,280
Speaker 1: between science and science fiction. Right. One aspect of science

205
00:09:49,360 --> 00:09:51,040
Speaker 1: is like, well, what are the laws? Could they be? This?

206
00:09:51,200 --> 00:09:51,680
Speaker 1: Could they be?

207
00:09:51,760 --> 00:09:51,920
Speaker 5: That?

208
00:09:51,920 --> 00:09:56,200
Speaker 1: There's an element of creativity and exploration there, absolutely, So, Yeah,

209
00:09:56,240 --> 00:09:59,640
Speaker 1: I'm constantly creating science fiction universes and trying to see

210
00:09:59,679 --> 00:10:00,559
Speaker 1: if they line.

211
00:10:00,440 --> 00:10:03,079
Speaker 3: Up with ours, and then you wonder why the other

212
00:10:03,120 --> 00:10:04,240
Speaker 3: physicis don't want to play with you.

213
00:10:06,320 --> 00:10:09,120
Speaker 1: Fortunately I got tenured before I revealed all of these

214
00:10:09,160 --> 00:10:10,959
Speaker 1: crazy instincts. That's the game.

215
00:10:11,280 --> 00:10:15,040
Speaker 3: Yeah, well, this is an interesting question, and so as usual,

216
00:10:15,120 --> 00:10:17,840
Speaker 3: we were wondering how many people out there had thought

217
00:10:17,880 --> 00:10:20,800
Speaker 3: about why scientists do the things they do, and in particular,

218
00:10:20,960 --> 00:10:23,280
Speaker 3: why they do simulations in their work.

219
00:10:23,640 --> 00:10:26,280
Speaker 1: So thanks very much to everybody who answers these questions

220
00:10:26,320 --> 00:10:29,199
Speaker 1: for this fun segment of the podcast, one of my favorites.

221
00:10:29,480 --> 00:10:32,080
Speaker 1: If you'd like to join the team or just answer

222
00:10:32,080 --> 00:10:34,920
Speaker 1: one or two questions right to us to questions at

223
00:10:35,040 --> 00:10:36,840
Speaker 1: Danielandjorge dot com.

224
00:10:37,000 --> 00:10:39,160
Speaker 3: So think about it for a second. If someone asks

225
00:10:39,200 --> 00:10:42,640
Speaker 3: you why scientists do simulations, what would you say.

226
00:10:42,600 --> 00:10:49,160
Speaker 6: Well, using simulations, we can observe scenarios in our models

227
00:10:49,679 --> 00:10:54,920
Speaker 6: that we can't necessarily observe in real life and see

228
00:10:55,280 --> 00:10:58,880
Speaker 6: what can happen in certain situations, like in a black

229
00:10:58,880 --> 00:11:01,320
Speaker 6: hole or when galaxies collide or something like that.

230
00:11:01,640 --> 00:11:07,080
Speaker 7: Scientists do simulations because the universe is really old, really big,

231
00:11:07,840 --> 00:11:11,959
Speaker 7: sometimes really destructive, Frankly, I'm happy they do a lot

232
00:11:12,000 --> 00:11:15,560
Speaker 7: of that modeling and simulations and don't necessarily try to

233
00:11:15,600 --> 00:11:20,320
Speaker 7: create Big Bang conditions on a big scale, or the

234
00:11:20,480 --> 00:11:22,040
Speaker 7: explosion of stars or something.

235
00:11:22,600 --> 00:11:25,720
Speaker 3: All Right, a couple of interesting answers. Did anyone look

236
00:11:25,760 --> 00:11:27,280
Speaker 3: at you funny when you ask them the question?

237
00:11:28,520 --> 00:11:30,640
Speaker 1: I don't know. These were all on the internet, so

238
00:11:30,800 --> 00:11:32,880
Speaker 1: I couldn't capture their facial expressions.

239
00:11:32,920 --> 00:11:34,040
Speaker 3: Did they send an emoji?

240
00:11:35,400 --> 00:11:38,600
Speaker 1: But I do sense some relief in there that, for example,

241
00:11:38,600 --> 00:11:41,360
Speaker 1: we are trying to simulate galaxy collisions rather than trying

242
00:11:41,360 --> 00:11:43,199
Speaker 1: to arrange galaxy collisions.

243
00:11:43,360 --> 00:11:45,160
Speaker 3: Well, if we could do that, that'd be pretty cool.

244
00:11:45,960 --> 00:11:48,400
Speaker 3: I mean, not for those galaxies, but just to have

245
00:11:48,440 --> 00:11:48,920
Speaker 3: that power.

246
00:11:49,120 --> 00:11:51,480
Speaker 1: Yeah, you'd have to have like sign offs from every

247
00:11:51,520 --> 00:11:55,319
Speaker 1: alien civilization in both galaxies before you can even begin.

248
00:11:55,559 --> 00:11:57,600
Speaker 3: I guess that would be the polite thing to do. Yes.

249
00:11:58,040 --> 00:11:59,720
Speaker 3: But anyways, as you were saying, this is a big

250
00:11:59,720 --> 00:12:03,480
Speaker 3: part of how science is done these days, and so, Daniel,

251
00:12:03,520 --> 00:12:05,760
Speaker 3: I guess let's start from the basics. What is a

252
00:12:05,800 --> 00:12:07,440
Speaker 3: simulation in your view?

253
00:12:07,400 --> 00:12:11,120
Speaker 1: Asy physicists, So, a simulation, or more specifically, a computer simulation,

254
00:12:11,559 --> 00:12:15,520
Speaker 1: is a specific program that involves a scientific model. A

255
00:12:15,640 --> 00:12:18,880
Speaker 1: model is like our picture of how the world might work.

256
00:12:19,160 --> 00:12:22,680
Speaker 1: It's like a simplified version of the real universe, one

257
00:12:22,679 --> 00:12:26,199
Speaker 1: that lets us explore a specific question. And a simulation

258
00:12:26,480 --> 00:12:29,520
Speaker 1: is usually a program on a computer that uses like

259
00:12:29,600 --> 00:12:33,760
Speaker 1: step by step methods to explore the behavior of that model.

260
00:12:34,600 --> 00:12:38,240
Speaker 3: And usually this model has the form of an equation, right, like,

261
00:12:38,280 --> 00:12:41,200
Speaker 3: for example, F EQUALSMA is a model of the world, right,

262
00:12:41,240 --> 00:12:42,640
Speaker 3: and how things move in the world.

263
00:12:42,800 --> 00:12:45,360
Speaker 1: Yeah, the science we do is mathematical, and the way

264
00:12:45,360 --> 00:12:49,440
Speaker 1: we describe things is mathematical, and so usually that involves equations,

265
00:12:49,679 --> 00:12:52,440
Speaker 1: equations that represent constraints on the model, like the way

266
00:12:52,520 --> 00:12:54,920
Speaker 1: things have to happen. And as you say, F equals

267
00:12:55,000 --> 00:12:56,800
Speaker 1: M is a model. If I want to toss a

268
00:12:56,840 --> 00:12:59,439
Speaker 1: baseball across my backyard, I want to answer the question

269
00:13:00,120 --> 00:13:02,360
Speaker 1: or is it going to land? Then I have lots

270
00:13:02,360 --> 00:13:05,120
Speaker 1: of possible ways to answer that question, but the most

271
00:13:05,120 --> 00:13:09,520
Speaker 1: appropriate ways to make the simplest model possible that still

272
00:13:09,559 --> 00:13:13,440
Speaker 1: captures everything that I'm interested in, And so often in

273
00:13:13,480 --> 00:13:16,040
Speaker 1: our world, like when we're tossing baseball's we can do

274
00:13:16,120 --> 00:13:19,480
Speaker 1: something pretty simple just F equals MA, which ignores all

275
00:13:19,520 --> 00:13:23,000
Speaker 1: sorts of swarming quantum details about what's happening inside the

276
00:13:23,040 --> 00:13:26,760
Speaker 1: baseball and just describes simple motion of a parabole.

277
00:13:27,200 --> 00:13:29,800
Speaker 3: Yeah, it's almost like you. I mean, as a scientist,

278
00:13:29,800 --> 00:13:31,920
Speaker 3: you're trying to come up with the rules of the universe, right,

279
00:13:31,960 --> 00:13:34,440
Speaker 3: that's sort of the goal of science, right, And what

280
00:13:34,679 --> 00:13:38,319
Speaker 3: sometimes that rule looks like is in a question that says,

281
00:13:38,360 --> 00:13:40,679
Speaker 3: you know, if you have a mass and you apply

282
00:13:40,720 --> 00:13:42,480
Speaker 3: a force to it, then it's going to start moving

283
00:13:42,480 --> 00:13:43,640
Speaker 3: with a certain acceleration.

284
00:13:43,800 --> 00:13:46,680
Speaker 1: Exactly in your words, these are all science fictions. We're

285
00:13:46,679 --> 00:13:48,920
Speaker 1: living in this world and we're wondering what are the rules,

286
00:13:49,000 --> 00:13:50,880
Speaker 1: and so we're trying a bunch of different rules, saying

287
00:13:50,880 --> 00:13:53,360
Speaker 1: does this rule describe our universe? Does that rule describe

288
00:13:53,400 --> 00:13:56,800
Speaker 1: our universe? So every sort of theoretical exploration of the

289
00:13:56,880 --> 00:14:00,000
Speaker 1: universe involves building a model and then asking the question

290
00:14:00,559 --> 00:14:03,839
Speaker 1: does that model align with the reality that we see.

291
00:14:03,920 --> 00:14:06,719
Speaker 1: Computer simulations are a special kind of model or a

292
00:14:06,720 --> 00:14:09,920
Speaker 1: special but A tests really complicated models that we can't

293
00:14:09,960 --> 00:14:13,000
Speaker 1: otherwise test, Like the model F equals M A pretty

294
00:14:13,040 --> 00:14:15,560
Speaker 1: simple I can use pencil and paper to make predictions,

295
00:14:15,840 --> 00:14:17,640
Speaker 1: and then I can throw a ball in my backyard

296
00:14:17,679 --> 00:14:18,959
Speaker 1: to confirm those predictions.

297
00:14:19,240 --> 00:14:21,400
Speaker 3: Right, because I guess F equals to A has like

298
00:14:21,440 --> 00:14:23,840
Speaker 3: a mathematical solution. But I think the idea is that

299
00:14:23,880 --> 00:14:25,480
Speaker 3: you take in a question like F equals to A

300
00:14:25,680 --> 00:14:28,480
Speaker 3: and you basically program that into a computer and say,

301
00:14:28,600 --> 00:14:31,760
Speaker 3: you know, any masses in this program, they have to

302
00:14:31,800 --> 00:14:33,240
Speaker 3: move according to this law.

303
00:14:33,480 --> 00:14:35,760
Speaker 1: That's right. If, for example, I don't just want to

304
00:14:35,800 --> 00:14:38,200
Speaker 1: describe one ball, but I want to describe like ten

305
00:14:38,240 --> 00:14:42,200
Speaker 1: to the twenty five balls, or some huge number of balls.

306
00:14:42,440 --> 00:14:45,360
Speaker 1: Maybe I'm modeling an ideal gas, or like a swimming

307
00:14:45,400 --> 00:14:47,920
Speaker 1: pool full of ping pong balls or something, and I

308
00:14:47,960 --> 00:14:49,800
Speaker 1: want to describe that. Then I can no longer use

309
00:14:49,840 --> 00:14:51,760
Speaker 1: pencil and paper. But you're in. I can take those

310
00:14:51,800 --> 00:14:54,040
Speaker 1: equations and put them into a computer and ask the

311
00:14:54,040 --> 00:14:56,920
Speaker 1: computer to force those balls to follow that equation, and

312
00:14:56,960 --> 00:14:59,520
Speaker 1: then I could see what happens. It's sort of like

313
00:14:59,600 --> 00:15:01,359
Speaker 1: a virtual experiment.

314
00:15:01,760 --> 00:15:04,360
Speaker 3: Yeah, it's like you're creating your own little universe.

315
00:15:04,000 --> 00:15:07,520
Speaker 1: Right exactly. And this becomes super essential when we don't

316
00:15:07,520 --> 00:15:10,880
Speaker 1: have like a single equation that describes everything, Like we

317
00:15:10,920 --> 00:15:14,080
Speaker 1: don't have a solution to what happens when you put

318
00:15:14,120 --> 00:15:16,600
Speaker 1: ten to the twenty five ping pong balls into a

319
00:15:16,600 --> 00:15:18,720
Speaker 1: swimming pool. We just don't know how to do that

320
00:15:18,800 --> 00:15:22,240
Speaker 1: calculation to come up with some nice summary of the results.

321
00:15:22,680 --> 00:15:24,200
Speaker 1: But what we can do is put it into a

322
00:15:24,240 --> 00:15:26,600
Speaker 1: computer and have the computer step it forward in time

323
00:15:27,000 --> 00:15:30,000
Speaker 1: very carefully, and we can see what happens without ever

324
00:15:30,080 --> 00:15:32,600
Speaker 1: actually having to buy that number of ping pong balls.

325
00:15:32,760 --> 00:15:35,520
Speaker 3: Right. That's sort of the power of the computer, right,

326
00:15:35,600 --> 00:15:39,240
Speaker 3: Like you can simulate one ball, which is a calculator, right,

327
00:15:39,320 --> 00:15:41,200
Speaker 3: Like you can say after one second, it's going to

328
00:15:41,200 --> 00:15:42,920
Speaker 3: be here, after two seconds, it's going to be here

329
00:15:43,800 --> 00:15:46,040
Speaker 3: by following these rules. But if you have, like you said,

330
00:15:46,080 --> 00:15:50,040
Speaker 3: a whole bunch of balls, or a more complicated system,

331
00:15:50,200 --> 00:15:52,680
Speaker 3: then a computer can sort of do all those calculations

332
00:15:52,760 --> 00:15:54,280
Speaker 3: for you faster exactly.

333
00:15:54,400 --> 00:15:57,760
Speaker 1: One huge advantage is tackling a very large number of objects,

334
00:15:57,920 --> 00:16:00,800
Speaker 1: and the other is when we don't have the equations,

335
00:16:00,840 --> 00:16:03,280
Speaker 1: we don't know how to solve them. Like for f equals,

336
00:16:03,280 --> 00:16:05,520
Speaker 1: I may we know how to solve that. Technically, that

337
00:16:05,720 --> 00:16:09,680
Speaker 1: is a differential equation because A is a second derivative

338
00:16:09,720 --> 00:16:13,000
Speaker 1: of position. Right, there's derivatives on both sides, and in

339
00:16:13,040 --> 00:16:16,200
Speaker 1: general and mathematics, differential equations are very very hard to solve.

340
00:16:16,200 --> 00:16:18,840
Speaker 1: This A small number that we actually know how to solve.

341
00:16:19,280 --> 00:16:21,640
Speaker 1: So sometimes you have a system that's described by a

342
00:16:21,680 --> 00:16:24,320
Speaker 1: differential equation and we don't know how to solve. Like

343
00:16:24,520 --> 00:16:27,720
Speaker 1: fluid flow, for example, described by the Navier Stokes equation,

344
00:16:27,920 --> 00:16:30,240
Speaker 1: we don't know how to solve that in general. But

345
00:16:30,360 --> 00:16:32,720
Speaker 1: what we can do on a computer is approximated. We

346
00:16:32,760 --> 00:16:34,960
Speaker 1: can say, you know, let's just move it forward in time,

347
00:16:35,280 --> 00:16:38,080
Speaker 1: not a year or a minute or some long period

348
00:16:38,120 --> 00:16:41,440
Speaker 1: of time, but just like a microsecond, and across a microsecond,

349
00:16:41,480 --> 00:16:43,880
Speaker 1: we can make some approximations. We can say, let's not

350
00:16:44,000 --> 00:16:46,200
Speaker 1: use the full equation, let's simplify it and take some

351
00:16:46,280 --> 00:16:48,600
Speaker 1: like linear approximation of it. And then if we take

352
00:16:48,600 --> 00:16:50,720
Speaker 1: a lot of tiny little steps, we hope that we

353
00:16:50,840 --> 00:16:52,240
Speaker 1: roughly get the right answer.

354
00:16:52,080 --> 00:16:54,040
Speaker 3: Right, because I think, as you were saying, like something

355
00:16:54,080 --> 00:16:57,080
Speaker 3: like f equos has the solution, meaning that you can

356
00:16:57,280 --> 00:17:00,080
Speaker 3: derive a formula for like the precision of your ball

357
00:17:00,080 --> 00:17:03,440
Speaker 3: at all times, where you can just like after three seconds,

358
00:17:03,520 --> 00:17:04,879
Speaker 3: you just put the time in and it gives you

359
00:17:04,920 --> 00:17:07,280
Speaker 3: the position of the ball, right, because you can integrate

360
00:17:07,320 --> 00:17:09,960
Speaker 3: that equation and find the solution. But some equations you can,

361
00:17:10,040 --> 00:17:12,720
Speaker 3: like they're so complex you can get a formally that

362
00:17:12,760 --> 00:17:14,560
Speaker 3: will tell you what's going to happen ten years from

363
00:17:14,600 --> 00:17:16,439
Speaker 3: now or twenty years from now, right those you need

364
00:17:16,480 --> 00:17:18,120
Speaker 3: to do little steps by little.

365
00:17:17,880 --> 00:17:20,680
Speaker 1: Steps, exactly. And the crucial idea there is that you're

366
00:17:20,720 --> 00:17:23,440
Speaker 1: making a linear approximation. You're taking the full equation which

367
00:17:23,440 --> 00:17:24,959
Speaker 1: you don't know how to solve, and you're saying, well,

368
00:17:25,000 --> 00:17:28,040
Speaker 1: let's replace it with an approximate version of it, which

369
00:17:28,080 --> 00:17:29,679
Speaker 1: is not going to be correct, but it might be

370
00:17:29,760 --> 00:17:32,760
Speaker 1: correct for like a microsecond. And so we'll use the

371
00:17:32,800 --> 00:17:35,200
Speaker 1: approximate linear version of that that we do know how

372
00:17:35,200 --> 00:17:37,440
Speaker 1: to solve for a tiny little step, and then we'll

373
00:17:37,440 --> 00:17:39,720
Speaker 1: start again, and we'll make another tiny little step, and

374
00:17:39,760 --> 00:17:42,160
Speaker 1: we hope them little mistakes cancel out and don't build

375
00:17:42,240 --> 00:17:44,040
Speaker 1: up into some big overall mistake.

376
00:17:44,480 --> 00:17:47,199
Speaker 3: Right yeah. It's almost like if you take small enough steps,

377
00:17:47,240 --> 00:17:50,800
Speaker 3: then you're less likely to deviate from the reality of.

378
00:17:50,760 --> 00:17:53,359
Speaker 1: It, right yeah, exactly. And people who do approximations know

379
00:17:53,440 --> 00:17:55,359
Speaker 1: that there's lots of times this is useful, Like you

380
00:17:55,400 --> 00:17:58,160
Speaker 1: want to calculate Trigg function like sign sign is really

381
00:17:58,200 --> 00:18:01,280
Speaker 1: hard to calculate like sort of from scratch, but for

382
00:18:01,440 --> 00:18:04,520
Speaker 1: very small values of the angle sign of x is

383
00:18:04,560 --> 00:18:07,440
Speaker 1: just equal to x. You can like approximate this complicated

384
00:18:07,480 --> 00:18:10,080
Speaker 1: function with a simple one. It mostly gets the right answer.

385
00:18:10,240 --> 00:18:11,280
Speaker 1: That's just one example.

386
00:18:11,359 --> 00:18:15,679
Speaker 3: I'm not sure going to trigonometry usually makes things to understand,

387
00:18:16,440 --> 00:18:18,639
Speaker 3: but I think we get the idea, which is that

388
00:18:18,840 --> 00:18:20,560
Speaker 3: you know, if you take small enough steps and you

389
00:18:21,920 --> 00:18:24,480
Speaker 3: sort of a simplify version of your model, then you're

390
00:18:24,560 --> 00:18:27,359
Speaker 3: less likely to make mistakes.

391
00:18:27,520 --> 00:18:30,600
Speaker 1: Exactly, you can't trust those approximations forward a second or

392
00:18:30,640 --> 00:18:32,600
Speaker 1: a minute or a year, but you could trust them

393
00:18:32,600 --> 00:18:35,920
Speaker 1: like a microsecond. And so you have the simulated universe

394
00:18:36,040 --> 00:18:39,119
Speaker 1: in your computer. You feed in the initial conditions, and

395
00:18:39,160 --> 00:18:41,120
Speaker 1: then you ask it to take a step forward in time,

396
00:18:41,400 --> 00:18:43,520
Speaker 1: and you ask it to take another step forward, and

397
00:18:43,520 --> 00:18:45,480
Speaker 1: if you have enough computing power, you can run it

398
00:18:45,520 --> 00:18:47,879
Speaker 1: for a while and you can see what happens to

399
00:18:47,880 --> 00:18:50,480
Speaker 1: all my ping pung balls in my simulated swimming pool,

400
00:18:50,800 --> 00:18:53,400
Speaker 1: or what happens to my galaxy as the stars all

401
00:18:53,440 --> 00:18:54,640
Speaker 1: swirl around each other.

402
00:18:54,800 --> 00:18:57,960
Speaker 3: Right, Like you were saying, like fluids are notoriously really

403
00:18:58,040 --> 00:19:01,400
Speaker 3: hard to solve as an equation, right, these are really

404
00:19:01,440 --> 00:19:04,680
Speaker 3: complex equations that govern what's going on because they sort

405
00:19:04,680 --> 00:19:06,440
Speaker 3: of like depend on a lot of things, Like there's

406
00:19:06,440 --> 00:19:08,840
Speaker 3: a lot going on, right, there's time, and then there's

407
00:19:08,880 --> 00:19:11,879
Speaker 3: distance and the velocity of things and all those factor

408
00:19:11,960 --> 00:19:15,280
Speaker 3: in that's hard or impossible to like predict exactly what's

409
00:19:15,280 --> 00:19:16,760
Speaker 3: going to happen in the future exactly.

410
00:19:16,840 --> 00:19:20,040
Speaker 1: And the big complication there is the interactions. Like back

411
00:19:20,080 --> 00:19:21,679
Speaker 1: to the ping pong balls. If you just had a

412
00:19:21,680 --> 00:19:23,800
Speaker 1: lot of ping pong balls and they're all flying around

413
00:19:23,840 --> 00:19:26,119
Speaker 1: and not touching each other, it wouldn't be that hard

414
00:19:26,160 --> 00:19:28,760
Speaker 1: to calculate what's going to happen to each one, But

415
00:19:28,800 --> 00:19:30,960
Speaker 1: as soon as they start banging against each other, becomes

416
00:19:31,040 --> 00:19:33,840
Speaker 1: much much more complicated because the solution of ping pong

417
00:19:33,880 --> 00:19:36,639
Speaker 1: ball number six hundred and forty two now depends on

418
00:19:36,760 --> 00:19:40,240
Speaker 1: ping pong ball number one hundred and eleven and every

419
00:19:40,240 --> 00:19:43,159
Speaker 1: other ping pong ball, so becomes much more complicated. And

420
00:19:43,200 --> 00:19:46,320
Speaker 1: that's why fluds are so complicated, because every like sheet

421
00:19:46,400 --> 00:19:48,520
Speaker 1: of the fluid depends on the friction with the other

422
00:19:48,560 --> 00:19:50,600
Speaker 1: sheet of the fluid. And that's what makes the knave

423
00:19:50,640 --> 00:19:52,480
Speaker 1: your Stokes equation for example, so.

424
00:19:52,560 --> 00:19:55,359
Speaker 3: Intractable, right, And so you take it little by little,

425
00:19:55,400 --> 00:19:57,000
Speaker 3: and so you say, Okay, at this time, I'm going

426
00:19:57,080 --> 00:19:59,560
Speaker 3: to ignore some of these effects and just take one

427
00:19:59,560 --> 00:20:02,200
Speaker 3: small star app to see where the all those little

428
00:20:02,240 --> 00:20:05,360
Speaker 3: molecules go. And then you keep repeating that and hopefully

429
00:20:05,880 --> 00:20:07,879
Speaker 3: it sort of looks like the real thing.

430
00:20:07,880 --> 00:20:10,919
Speaker 1: Exactly, and it gives you this incredible power that you

431
00:20:10,960 --> 00:20:15,040
Speaker 1: can hopefully identify emergent behavior. The way we do science

432
00:20:15,080 --> 00:20:17,360
Speaker 1: in our universe is that we like focus on one

433
00:20:17,480 --> 00:20:20,199
Speaker 1: level where we understand things we can describe, like the

434
00:20:20,240 --> 00:20:23,640
Speaker 1: microphysics of how particles being against each other. But sometimes

435
00:20:23,680 --> 00:20:26,320
Speaker 1: we're interested in things at another level, like you understand

436
00:20:26,320 --> 00:20:28,440
Speaker 1: how rain drops move through the wind, but your real

437
00:20:28,520 --> 00:20:31,200
Speaker 1: question is like is this hurricane going to hit Florida

438
00:20:31,320 --> 00:20:33,679
Speaker 1: or Alabama? And so even if you don't have like

439
00:20:33,680 --> 00:20:36,399
Speaker 1: an equation that describes hurricanes, if you have an equation

440
00:20:36,480 --> 00:20:39,159
Speaker 1: that describes the rain drops, you can feed that all

441
00:20:39,200 --> 00:20:42,240
Speaker 1: into your computer, run simulations, and then get answers to

442
00:20:42,320 --> 00:20:45,400
Speaker 1: your higher level question. You can see like the emergent

443
00:20:45,440 --> 00:20:48,000
Speaker 1: phenomena of the hurricane in simulation.

444
00:20:48,280 --> 00:20:51,080
Speaker 3: Well, it's dig a little bit deeper into how simulation

445
00:20:51,200 --> 00:20:55,240
Speaker 3: works and the things like weather and why scientists use

446
00:20:55,240 --> 00:20:58,800
Speaker 3: simulations to try to learn things about the real universe.

447
00:20:59,040 --> 00:21:00,600
Speaker 3: But first let's take a click break.

448
00:21:04,520 --> 00:21:07,479
Speaker 1: With big wireless providers, what you see is never what

449
00:21:07,560 --> 00:21:10,240
Speaker 1: you get. Somewhere between the store and your first month's bill,

450
00:21:10,280 --> 00:21:14,000
Speaker 1: the price, your thoughts you were paying magically skyrockets. With Mintmobile,

451
00:21:14,119 --> 00:21:17,120
Speaker 1: You'll never have to worry about gotcha's ever again. When

452
00:21:17,160 --> 00:21:19,400
Speaker 1: Mint Mobile says fifteen dollars a month for a three

453
00:21:19,440 --> 00:21:22,439
Speaker 1: month plan, they really mean it. I've used Mintmobile and

454
00:21:22,480 --> 00:21:25,160
Speaker 1: the call quality is always so crisp and so clear

455
00:21:25,240 --> 00:21:27,400
Speaker 1: I can recommend it to you, So say bye bye

456
00:21:27,440 --> 00:21:30,720
Speaker 1: to your overpriced wireless plans, jaw dropping monthly bills and

457
00:21:30,840 --> 00:21:33,920
Speaker 1: unexpected overages. You can use your own phone with any

458
00:21:33,960 --> 00:21:36,600
Speaker 1: mint Mobile plan and bring your phone number along with

459
00:21:36,680 --> 00:21:39,720
Speaker 1: your existing contacts. So ditch your overpriced wireless with mint

460
00:21:39,720 --> 00:21:42,600
Speaker 1: Mobiles deal and get three months a premium wireless service

461
00:21:42,600 --> 00:21:44,800
Speaker 1: for fifteen bucks a month. To get this new customer

462
00:21:44,880 --> 00:21:47,040
Speaker 1: offer and your new three month premium wireless plan for

463
00:21:47,119 --> 00:21:49,280
Speaker 1: just fifteen bucks a month, go to mintmobile dot com

464
00:21:49,280 --> 00:21:52,600
Speaker 1: slash universe. That's mintmobile dot com slash universe. Cut your

465
00:21:52,600 --> 00:21:54,919
Speaker 1: wireless bill to fifteen bucks a month. At mintmobile dot

466
00:21:54,960 --> 00:21:58,280
Speaker 1: com slash universe. Forty five dollars upfront payment required equivalent

467
00:21:58,320 --> 00:22:00,680
Speaker 1: to fifteen dollars per month new customer on first three

468
00:22:00,720 --> 00:22:03,719
Speaker 1: month plan only speeds slower about forty gigabytes on unlimited plan.

469
00:22:03,800 --> 00:22:07,080
Speaker 1: Additional taxi spees and restrictions apply. Seement Mobile for details.

470
00:22:07,240 --> 00:22:10,280
Speaker 8: AI might be the most important new computer technology ever.

471
00:22:10,560 --> 00:22:13,760
Speaker 8: It's storming every industry and literally billions of dollars are

472
00:22:13,800 --> 00:22:18,000
Speaker 8: being invested, so buckle up. The problem is that AI

473
00:22:18,119 --> 00:22:20,960
Speaker 8: needs a lot of speed and processing power, So how

474
00:22:20,960 --> 00:22:24,120
Speaker 8: do you compete without cost spiraling out of control. It's

475
00:22:24,160 --> 00:22:26,439
Speaker 8: time to upgrade to the next generation of the cloud.

476
00:22:26,800 --> 00:22:32,000
Speaker 8: Oracle Cloud Infrastructure or OCI. OCI is a single platform

477
00:22:32,040 --> 00:22:36,879
Speaker 8: for your infrastructure, database, application development, and AI needs. OCI

478
00:22:37,040 --> 00:22:40,159
Speaker 8: has forty eight times the bandwidth of other clouds, offers

479
00:22:40,160 --> 00:22:43,520
Speaker 8: one consistent price instead of variable regional pricing, and of

480
00:22:43,560 --> 00:22:46,639
Speaker 8: course nobody does data better than Oracle. So now you

481
00:22:46,640 --> 00:22:49,040
Speaker 8: can train your AI models at twice the speed and

482
00:22:49,160 --> 00:22:51,399
Speaker 8: less than half the cost of other clouds. If you

483
00:22:51,440 --> 00:22:53,880
Speaker 8: want to do more and spend less, like Uber eight

484
00:22:53,880 --> 00:22:56,560
Speaker 8: by eight and Data Bricks Mosaic, take a free test

485
00:22:56,600 --> 00:23:00,800
Speaker 8: drive of OCI at Oracle dot com slash strateg that's

486
00:23:00,840 --> 00:23:05,840
Speaker 8: Oracle dot com slash Strategic Oracle dot Com Slash Strategic.

487
00:23:06,119 --> 00:23:09,000
Speaker 1: If you love iPhone, you'll love Apple Card. It's the

488
00:23:09,080 --> 00:23:12,959
Speaker 1: credit card designed for iPhone. It gives you unlimited daily

489
00:23:13,080 --> 00:23:16,040
Speaker 1: cash back that can earn four point four zero percent

490
00:23:16,200 --> 00:23:19,000
Speaker 1: annual percentage yield. When you open a high Yield savings

491
00:23:19,040 --> 00:23:22,560
Speaker 1: account through Applecard, apply for Applecard in the wallet app

492
00:23:22,680 --> 00:23:26,200
Speaker 1: subject to credit approval. Savings is available to Applecard owners

493
00:23:26,240 --> 00:23:29,480
Speaker 1: subject to eligibility. Apple Card and Savings by Goldman Sachs

494
00:23:29,520 --> 00:23:32,920
Speaker 1: Bank USA, Salt Lake City Branch Member FDIC terms and

495
00:23:33,040 --> 00:23:35,600
Speaker 1: more at applecard dot com. When you pop a piece

496
00:23:35,600 --> 00:23:38,119
Speaker 1: of cheese into your mouth or enjoy a rich spoonful

497
00:23:38,160 --> 00:23:41,640
Speaker 1: of Greek yogurt, you're probably not thinking about the environmental

498
00:23:41,680 --> 00:23:44,520
Speaker 1: impact of each and every bite, But the people in

499
00:23:44,560 --> 00:23:47,520
Speaker 1: the dairy industry are US. Dairy has set themselves some

500
00:23:47,640 --> 00:23:52,600
Speaker 1: ambitious sustainability goals, including being greenhouse gas neutral by twenty fifty.

501
00:23:52,680 --> 00:23:54,760
Speaker 1: That's why they're working hard every day to find new

502
00:23:54,800 --> 00:23:58,080
Speaker 1: ways to reduce waste, conserve natural resources, and drive down

503
00:23:58,160 --> 00:24:01,359
Speaker 1: greenhouse gas emissions. Take a walk. For example, most dairy

504
00:24:01,400 --> 00:24:04,480
Speaker 1: farms reuse water up to four times the same water

505
00:24:04,600 --> 00:24:08,080
Speaker 1: cools the milk, cleans equipment, washes the barn and irrigates

506
00:24:08,080 --> 00:24:11,240
Speaker 1: the crops. How is US dairy tackling greenhouse gases? Many

507
00:24:11,320 --> 00:24:14,399
Speaker 1: farms use anaerobic digestors that turn the methane from maneuver

508
00:24:14,520 --> 00:24:18,320
Speaker 1: into renewable energy that can power farms, towns, and electric cars.

509
00:24:18,359 --> 00:24:20,119
Speaker 1: So the next time you grab a slice of pizza

510
00:24:20,200 --> 00:24:22,440
Speaker 1: or lick an ice cream cone, know that dairy farmers

511
00:24:22,440 --> 00:24:25,439
Speaker 1: and processors around the country are using the latest practices

512
00:24:25,480 --> 00:24:28,680
Speaker 1: and innovations to provide the nutrient dense dairy products we

513
00:24:28,800 --> 00:24:31,680
Speaker 1: love with less of an impact. Visit usdairy dot com

514
00:24:31,720 --> 00:24:33,520
Speaker 1: slash sustainability to learn more.

515
00:24:42,359 --> 00:24:45,000
Speaker 3: All Right, we're having a simulation of a podcast here, right,

516
00:24:45,040 --> 00:24:47,840
Speaker 3: We're not really having a podcast, right, We're just pretending

517
00:24:47,880 --> 00:24:48,680
Speaker 3: to have a podcast.

518
00:24:48,800 --> 00:24:52,640
Speaker 1: We're simulating the process of injecting ideas into listener minds.

519
00:24:52,840 --> 00:24:55,919
Speaker 3: And so we're talking about why scientists use simulations, and

520
00:24:55,960 --> 00:24:59,400
Speaker 3: it's kind of, I guess the philosophical question, perhaps because

521
00:24:59,520 --> 00:25:02,760
Speaker 3: doing a relation of reality is not really reality, right,

522
00:25:03,400 --> 00:25:07,320
Speaker 3: and you're not really experimenting on reality. So it's kind

523
00:25:07,320 --> 00:25:09,320
Speaker 3: of a I guess, a funny thing for scientists to

524
00:25:09,359 --> 00:25:11,600
Speaker 3: do it because you're not really doing experiments in the

525
00:25:11,640 --> 00:25:14,040
Speaker 3: real world. But it's at the same time really helpful.

526
00:25:14,119 --> 00:25:16,679
Speaker 1: Right, that's true, But that same criticism could be applied

527
00:25:16,680 --> 00:25:20,119
Speaker 1: to basically everything in science. When we do science, we

528
00:25:20,200 --> 00:25:23,560
Speaker 1: never use all of the full gory details of the

529
00:25:23,680 --> 00:25:27,400
Speaker 1: universe to answer a question. We're always using some stripped

530
00:25:27,400 --> 00:25:31,199
Speaker 1: down version because otherwise it's totally intractable. Like when we

531
00:25:31,240 --> 00:25:33,760
Speaker 1: do F equals M, even for a single ball flying

532
00:25:33,760 --> 00:25:36,480
Speaker 1: through the air, we're ignoring lots of stuff. We're ignoring

533
00:25:36,520 --> 00:25:39,920
Speaker 1: air resistance, we're ignoring quantum effects of the particles inside

534
00:25:39,920 --> 00:25:42,359
Speaker 1: of it. We're ignoring all sorts of things because we

535
00:25:42,400 --> 00:25:45,000
Speaker 1: don't think that they are important. And so every time

536
00:25:45,040 --> 00:25:48,040
Speaker 1: you build a model of the universe, theoretical or simulation,

537
00:25:48,400 --> 00:25:50,399
Speaker 1: you're always making a choice about what to ignore and

538
00:25:50,440 --> 00:25:51,159
Speaker 1: what to include.

539
00:25:51,440 --> 00:25:54,080
Speaker 3: Well, we talked a lot about what a simulation is, right.

540
00:25:54,119 --> 00:25:56,480
Speaker 3: It's a computer program where you program in the rules

541
00:25:56,520 --> 00:25:59,639
Speaker 3: that you think that the world follows the rules of

542
00:25:59,640 --> 00:26:01,879
Speaker 3: the universe, at least in your simulated universe, and then

543
00:26:01,920 --> 00:26:05,199
Speaker 3: you sort of let the computer kind of run this world,

544
00:26:05,400 --> 00:26:07,440
Speaker 3: and then it sort of tells you what may or

545
00:26:07,480 --> 00:26:09,480
Speaker 3: may will sort of happen.

546
00:26:09,359 --> 00:26:12,240
Speaker 1: Exactly, and it lets you examine all sorts of universes

547
00:26:12,280 --> 00:26:15,080
Speaker 1: you don't otherwise have access to, like in my work,

548
00:26:15,080 --> 00:26:17,320
Speaker 1: and lets me answer questions like what would I see

549
00:26:17,359 --> 00:26:20,840
Speaker 1: in our particle detectors if the higgs boson was this

550
00:26:20,960 --> 00:26:23,560
Speaker 1: kind of particle, or what if there was no Higgs boson,

551
00:26:24,040 --> 00:26:26,080
Speaker 1: or what if it had twice the mass that it had?

552
00:26:26,160 --> 00:26:28,639
Speaker 1: What would we see in our detectors? What would that

553
00:26:28,720 --> 00:26:31,520
Speaker 1: universe be like? What would those experiments result in?

554
00:26:31,800 --> 00:26:34,440
Speaker 3: So it gives you ideas for experiments, or it's a

555
00:26:34,480 --> 00:26:37,400
Speaker 3: sort it can guide your real experiments. Right, that's part

556
00:26:37,440 --> 00:26:37,879
Speaker 3: of the idea.

557
00:26:37,960 --> 00:26:41,760
Speaker 1: Right, it's actually crucial for interpreting our experiments. When we

558
00:26:41,800 --> 00:26:43,920
Speaker 1: look at data from the actual collider and we see

559
00:26:43,960 --> 00:26:46,479
Speaker 1: these splashes of energy here and splashes of energy there,

560
00:26:46,520 --> 00:26:49,080
Speaker 1: and we look at the patterns the correlations. The way

561
00:26:49,119 --> 00:26:53,080
Speaker 1: we interpret those is by comparing them to simulations. We say,

562
00:26:53,640 --> 00:26:56,320
Speaker 1: is this consistent with the higgs boson with these properties

563
00:26:56,400 --> 00:26:58,159
Speaker 1: or is it more consistent with the higgs boson with

564
00:26:58,200 --> 00:27:02,320
Speaker 1: some other properties. Simulation in some sense defines the ideas

565
00:27:02,320 --> 00:27:05,040
Speaker 1: that we're considering, the various hypotheses that we're trying to

566
00:27:05,080 --> 00:27:05,879
Speaker 1: distinguish between.

567
00:27:06,080 --> 00:27:09,040
Speaker 3: Right it, let's you explore the possibilities. That's the idea

568
00:27:09,040 --> 00:27:11,280
Speaker 3: of a simulation, right, Let's you maybe make mistakes.

569
00:27:11,359 --> 00:27:14,119
Speaker 1: Absolutely, and before we build a detector, we simulated to

570
00:27:14,119 --> 00:27:16,800
Speaker 1: see like is this going to work or how well

571
00:27:16,880 --> 00:27:19,200
Speaker 1: is it going to perform? Or oops, turns out we

572
00:27:19,320 --> 00:27:21,359
Speaker 1: need to swap the order these two things or nothing's

573
00:27:21,400 --> 00:27:23,720
Speaker 1: going to work. So yeah, making mistakes and simulation is

574
00:27:23,800 --> 00:27:25,640
Speaker 1: much cheaper than making them in reality.

575
00:27:25,800 --> 00:27:27,960
Speaker 3: Yeah, And as you were saying, simulations play a big

576
00:27:28,000 --> 00:27:30,639
Speaker 3: part in weather prediction, right, I mean that's how weather

577
00:27:30,720 --> 00:27:33,040
Speaker 3: predictions work. Like when you look at the weather forecast

578
00:27:33,280 --> 00:27:35,520
Speaker 3: and says it's going to rain tomorrow, it's because some

579
00:27:35,600 --> 00:27:38,520
Speaker 3: big computer out there has basically taken the data from

580
00:27:38,520 --> 00:27:41,760
Speaker 3: today and simulated what's going to happen tomorrow exactly.

581
00:27:41,840 --> 00:27:46,240
Speaker 1: And that's really the origin of computer simulations. People wanted

582
00:27:46,280 --> 00:27:49,120
Speaker 1: to predict the weather, to understand what's going to happen

583
00:27:49,160 --> 00:27:51,520
Speaker 1: to these cloud patterns, but nobody could really do it

584
00:27:51,560 --> 00:27:54,600
Speaker 1: with pencil and papers. Too complicated, too many pieces of information,

585
00:27:54,720 --> 00:27:57,840
Speaker 1: and the equations are really just a mess. So meteorology

586
00:27:57,920 --> 00:27:59,960
Speaker 1: is one of the first places where people decided, let's

587
00:28:00,119 --> 00:28:02,320
Speaker 1: code this up on the computer to try to grapple

588
00:28:02,400 --> 00:28:04,800
Speaker 1: with this complexity and see if we can get anything

589
00:28:04,920 --> 00:28:07,040
Speaker 1: right now. Just after World War Two when computers were

590
00:28:07,040 --> 00:28:10,800
Speaker 1: first displaying like computational power, and it was the weather

591
00:28:10,840 --> 00:28:14,280
Speaker 1: forecasters and the nuclear physicists that first really jumped on

592
00:28:14,280 --> 00:28:14,720
Speaker 1: this train.

593
00:28:15,240 --> 00:28:17,280
Speaker 3: Yeah, because I think the way the weather works is

594
00:28:17,280 --> 00:28:22,040
Speaker 3: that you have all this data, mateiological data, weather data

595
00:28:22,600 --> 00:28:25,399
Speaker 3: across let's say the United States, that tells you the

596
00:28:25,440 --> 00:28:28,040
Speaker 3: wind speeds and the clouds and the pressures and all that,

597
00:28:28,520 --> 00:28:30,200
Speaker 3: and then you can use it and put it into

598
00:28:30,800 --> 00:28:33,840
Speaker 3: basically your computer, which has a model of what should

599
00:28:33,840 --> 00:28:36,960
Speaker 3: happen next. If that's you have all these pressures and

600
00:28:37,000 --> 00:28:38,680
Speaker 3: wind patterns exactly.

601
00:28:38,920 --> 00:28:42,080
Speaker 1: And those models are not perfect, they don't describe everything,

602
00:28:42,320 --> 00:28:45,560
Speaker 1: so they're most reliable over short times because that's when

603
00:28:45,560 --> 00:28:47,440
Speaker 1: the errors are not going to compound as much, which

604
00:28:47,440 --> 00:28:49,160
Speaker 1: is why I like the prediction for how hot it's

605
00:28:49,160 --> 00:28:51,440
Speaker 1: going to be tomorrow is much more reliable than the

606
00:28:51,480 --> 00:28:54,480
Speaker 1: prediction for how hot it's going to be in ten years,

607
00:28:54,520 --> 00:28:57,120
Speaker 1: which you basically have no information about. Or if you

608
00:28:57,160 --> 00:28:59,640
Speaker 1: look at those projections for like where's the hurricane going

609
00:28:59,680 --> 00:29:03,120
Speaker 1: to be, the potential path of the hurricane gets wider

610
00:29:03,400 --> 00:29:06,200
Speaker 1: as the prediction gets further out because there's more uncertainty,

611
00:29:06,280 --> 00:29:08,320
Speaker 1: like is it going to hit Alabama? We don't know.

612
00:29:08,640 --> 00:29:10,720
Speaker 3: Yeah, it's pretty cool and actually an interesting fact. I

613
00:29:10,760 --> 00:29:14,240
Speaker 3: would just talk to a hurricane scientist a couple of

614
00:29:14,320 --> 00:29:16,959
Speaker 3: months ago, and he was saying that we're still at

615
00:29:17,000 --> 00:29:22,840
Speaker 3: the point apparently where humans outperform simulations, even supercomputer simulations.

616
00:29:22,400 --> 00:29:25,000
Speaker 1: Humans using pencil and paper, or just humans like with

617
00:29:25,040 --> 00:29:26,160
Speaker 1: their intuition.

618
00:29:25,920 --> 00:29:28,680
Speaker 3: Humans with their intuition. So like, apparently we're still at

619
00:29:28,720 --> 00:29:31,480
Speaker 3: the point where if if you're seeing a hurricane move,

620
00:29:31,760 --> 00:29:33,760
Speaker 3: you run a computer simulation about where it's going to

621
00:29:33,840 --> 00:29:37,120
Speaker 3: go next. A human or like a season experienced hurricane

622
00:29:37,160 --> 00:29:40,520
Speaker 3: watcher will still today better at predicting what the hurricane

623
00:29:40,560 --> 00:29:42,160
Speaker 3: is going to do, just from like what's going on

624
00:29:42,200 --> 00:29:45,120
Speaker 3: inside their brain and the history of what they've seen before.

625
00:29:45,240 --> 00:29:48,080
Speaker 3: But it's getting apparently closer and closer. So maybe in

626
00:29:48,080 --> 00:29:52,960
Speaker 3: the near future, computers will make those hurricane watchers totally obsolete.

627
00:29:53,000 --> 00:29:55,200
Speaker 1: That's super fascinating and it's fun to think about, like

628
00:29:55,280 --> 00:29:58,760
Speaker 1: what's going on inside that person's brain. They have built

629
00:29:58,760 --> 00:30:03,120
Speaker 1: in their head some neural network with literal biological neurons, right,

630
00:30:03,160 --> 00:30:06,360
Speaker 1: and not your typical artificial neural network that models hurricanes

631
00:30:06,400 --> 00:30:08,440
Speaker 1: and they've trained it on a bunch of real hurricanes,

632
00:30:08,720 --> 00:30:09,680
Speaker 1: So that's pretty cool.

633
00:30:09,920 --> 00:30:12,560
Speaker 3: Yeah, it makes you wonder if maybe, like in the future,

634
00:30:12,560 --> 00:30:15,680
Speaker 3: they're going to use AIS to predict the weather, maybe

635
00:30:15,720 --> 00:30:18,000
Speaker 3: you don't need a scientific model of what's going on.

636
00:30:18,160 --> 00:30:21,480
Speaker 1: That's a really fascinating question because AIS are already being

637
00:30:21,600 --> 00:30:25,760
Speaker 1: used to help boost simulations. One problem with simulations is

638
00:30:25,760 --> 00:30:29,360
Speaker 1: that they can be very expensive computationally. You have lots

639
00:30:29,360 --> 00:30:31,040
Speaker 1: and lots of rain jobs and you want to model

640
00:30:31,080 --> 00:30:33,680
Speaker 1: it very, very accurately. It takes a computer a long

641
00:30:33,720 --> 00:30:36,200
Speaker 1: time to calculate every rain job and move it forward

642
00:30:36,240 --> 00:30:38,760
Speaker 1: in time. You want to predict something a few days out,

643
00:30:38,920 --> 00:30:41,520
Speaker 1: it can be very expensive computationally. We run into this

644
00:30:41,560 --> 00:30:43,560
Speaker 1: problem in particle physics all the time because we want

645
00:30:43,560 --> 00:30:46,360
Speaker 1: to simulate billions and billions of potential collisions, and the

646
00:30:46,360 --> 00:30:49,800
Speaker 1: interactions with the detector are very complicated. So to generate

647
00:30:49,920 --> 00:30:53,600
Speaker 1: one simulated collision, for example, takes like thirty minutes even

648
00:30:53,640 --> 00:30:56,440
Speaker 1: on a modern computer. We use AI to boost those

649
00:30:56,480 --> 00:31:00,640
Speaker 1: to make them faster. Essentially, we train machine learning algorithms

650
00:31:00,680 --> 00:31:04,560
Speaker 1: to reproduce what the careful calculations have done. They don't

651
00:31:04,640 --> 00:31:07,520
Speaker 1: understand it. They don't like have the same equations built in.

652
00:31:07,600 --> 00:31:10,080
Speaker 1: It's just sort of like those people watching the examples

653
00:31:10,160 --> 00:31:13,520
Speaker 1: and getting an intuition. This is like a machine learning intuition.

654
00:31:14,640 --> 00:31:18,680
Speaker 3: So now you're not just similar working in a Meida world.

655
00:31:18,840 --> 00:31:21,200
Speaker 3: Now you're in twitting your way through a Meida world.

656
00:31:21,280 --> 00:31:23,440
Speaker 1: Yeah. And one problem is that we don't always know

657
00:31:23,600 --> 00:31:26,200
Speaker 1: if their predictions are accurate or why they make them.

658
00:31:26,200 --> 00:31:28,360
Speaker 1: You can't ask them like, hm, why did this go

659
00:31:28,600 --> 00:31:31,400
Speaker 1: left instead of right? They just have an internal model,

660
00:31:31,440 --> 00:31:34,400
Speaker 1: the same way your hurricane watchers probably can't answer detailed

661
00:31:34,440 --> 00:31:37,080
Speaker 1: questions about why they feel it's going this way. They

662
00:31:37,160 --> 00:31:37,760
Speaker 1: just feel it.

663
00:31:38,040 --> 00:31:41,920
Speaker 3: Yeah. And so it also raises these interesting philosophical questions

664
00:31:41,960 --> 00:31:44,640
Speaker 3: about what science is right, Like, is it still science

665
00:31:44,640 --> 00:31:46,520
Speaker 3: if you get an AI to predict what's going on,

666
00:31:46,720 --> 00:31:48,600
Speaker 3: even if you don't understand what the AI did.

667
00:31:48,760 --> 00:31:51,360
Speaker 1: It's a deep question that we're struggling with all the time.

668
00:31:51,480 --> 00:31:54,240
Speaker 1: But what's not controversial is that it gives us extraordinary

669
00:31:54,320 --> 00:31:57,200
Speaker 1: power to do things we just couldn't do otherwise. We

670
00:31:57,200 --> 00:31:59,760
Speaker 1: can now run our simulations for much much longer and

671
00:31:59,800 --> 00:32:01,880
Speaker 1: in much more depth. You want to know what's going

672
00:32:01,920 --> 00:32:04,360
Speaker 1: to happen when the Milky Way collides with Andromeda or

673
00:32:04,360 --> 00:32:08,240
Speaker 1: the far future of our universe. Simulations give you that power.

674
00:32:08,480 --> 00:32:10,360
Speaker 1: You don't have to sit around and wait for the

675
00:32:10,480 --> 00:32:13,200
Speaker 1: events to play out. We can test it in simulation.

676
00:32:13,720 --> 00:32:16,360
Speaker 3: Right. That's pretty cool. And so what are the different

677
00:32:16,400 --> 00:32:18,200
Speaker 3: types of simulations that scientists use.

678
00:32:18,600 --> 00:32:21,440
Speaker 1: I would say that the simulation is almost everywhere in science.

679
00:32:21,480 --> 00:32:23,480
Speaker 1: You know, it used to be limited to a few

680
00:32:23,520 --> 00:32:28,000
Speaker 1: computationally complex fields, but now everybody sees how useful it is.

681
00:32:28,520 --> 00:32:31,000
Speaker 1: You know, even big companies like you want to design

682
00:32:31,040 --> 00:32:33,760
Speaker 1: a new airplane and you're considering a few different wing shapes.

683
00:32:34,000 --> 00:32:36,360
Speaker 1: It used to be you have to build prototypes of

684
00:32:36,360 --> 00:32:38,719
Speaker 1: those wing shapes and put them in a real huge

685
00:32:38,800 --> 00:32:42,360
Speaker 1: wind tunnel, very time consuming expensive. Now you can just

686
00:32:42,520 --> 00:32:45,360
Speaker 1: simulate the wind tunnel and get an idea for which

687
00:32:45,400 --> 00:32:47,760
Speaker 1: wing shape is going to work. You can explore thousands

688
00:32:47,800 --> 00:32:51,880
Speaker 1: of different shapes simultaneously, so that can be very very powerful.

689
00:32:52,040 --> 00:32:55,080
Speaker 1: So I think simulations are essentially everywhere in science now.

690
00:32:55,240 --> 00:32:57,680
Speaker 3: Well, I think they've been using simulations in things like

691
00:32:57,720 --> 00:33:00,600
Speaker 3: aerospace for a long time, right, Like even I'm thinking

692
00:33:00,640 --> 00:33:03,080
Speaker 3: in the space program, in the fifties and sixties. I mean,

693
00:33:03,120 --> 00:33:06,440
Speaker 3: they didn't use physical computers, but they use people computers

694
00:33:06,480 --> 00:33:09,520
Speaker 3: to sort of simulate what the trajectories of the spacecraft

695
00:33:09,760 --> 00:33:10,280
Speaker 3: were going to be.

696
00:33:10,360 --> 00:33:13,680
Speaker 1: Right, They definitely used human brains to do those calculations.

697
00:33:14,120 --> 00:33:16,560
Speaker 1: Whether you consider that a simulation, I think is a

698
00:33:16,600 --> 00:33:20,360
Speaker 1: tricky point. Is that just a theoretical calculation which people

699
00:33:20,400 --> 00:33:23,080
Speaker 1: have been doing, you know since Galileo or Francis Bacon

700
00:33:23,200 --> 00:33:26,479
Speaker 1: or whatever. Is it actually a simulation? I don't know that.

701
00:33:26,480 --> 00:33:28,440
Speaker 3: That's a tough question, all right, So then what are

702
00:33:28,440 --> 00:33:31,240
Speaker 3: some of the other types of simulations people do, or

703
00:33:31,520 --> 00:33:34,000
Speaker 3: what are some other ways that physicists use simulations.

704
00:33:34,120 --> 00:33:36,880
Speaker 1: Another way they use them is to observe things that

705
00:33:36,920 --> 00:33:39,920
Speaker 1: they otherwise couldn't see. Like we want to know what's

706
00:33:39,920 --> 00:33:42,440
Speaker 1: going on inside the sun. Well, we have really no

707
00:33:42,640 --> 00:33:45,840
Speaker 1: prospects for actually seeing what's going on inside the sun,

708
00:33:46,320 --> 00:33:48,760
Speaker 1: but we can build a simulation of the inside of

709
00:33:48,760 --> 00:33:51,280
Speaker 1: the Sun, and that's going to make predictions for things

710
00:33:51,280 --> 00:33:53,640
Speaker 1: that we can see, things happening on the surface of

711
00:33:53,680 --> 00:33:56,120
Speaker 1: the Sun, or the number of neutrinos coming to Earth,

712
00:33:56,200 --> 00:33:58,600
Speaker 1: and that helps us get an understanding for what's really

713
00:33:58,640 --> 00:34:02,520
Speaker 1: happening inside the Sun, and in the simulation, you're not limited, right,

714
00:34:02,560 --> 00:34:04,880
Speaker 1: you can ask questions about anything that's happening, like what

715
00:34:04,960 --> 00:34:06,720
Speaker 1: is the temperature at the core of the Sun, what

716
00:34:06,920 --> 00:34:10,200
Speaker 1: is the velocity of the plasma. So often simulations always

717
00:34:10,280 --> 00:34:12,960
Speaker 1: checked by real experiments in places where we can observe

718
00:34:13,000 --> 00:34:16,240
Speaker 1: them give us access to things that we can't otherwise observe.

719
00:34:16,440 --> 00:34:19,120
Speaker 3: I think that's a crucial step in this process, right,

720
00:34:19,160 --> 00:34:21,360
Speaker 3: Like you can come up with this imaginary world in

721
00:34:21,400 --> 00:34:23,680
Speaker 3: your computer, but it has to match sort of what

722
00:34:23,760 --> 00:34:26,439
Speaker 3: you see at the end with reality.

723
00:34:26,120 --> 00:34:28,839
Speaker 1: Right, absolutely. Otherwise it's just science fiction, which you know

724
00:34:29,000 --> 00:34:32,239
Speaker 1: has its own value. But there's a special interest in

725
00:34:32,280 --> 00:34:35,040
Speaker 1: our universe and that's led to all sorts of deep understanding.

726
00:34:35,040 --> 00:34:37,360
Speaker 1: You know, the original simulations of the Sun predicted a

727
00:34:37,400 --> 00:34:39,960
Speaker 1: huge number of neutrinos landing on the surface of the Earth,

728
00:34:40,000 --> 00:34:42,040
Speaker 1: and they went out and measured them and the answer

729
00:34:42,160 --> 00:34:44,520
Speaker 1: was wrong. And they thought, did we get the Sun wrong?

730
00:34:44,640 --> 00:34:47,000
Speaker 1: Or is there something going on with neutrinos? And it

731
00:34:47,040 --> 00:34:49,120
Speaker 1: turned out the simulation of the Sun was correct and

732
00:34:49,200 --> 00:34:52,680
Speaker 1: neutrinos were doing something wonky between there and here, right.

733
00:34:52,719 --> 00:34:54,080
Speaker 3: I think the idea is that you know, if you

734
00:34:54,280 --> 00:34:57,399
Speaker 3: create a simulation, and you tweak the parameters of it, right,

735
00:34:57,520 --> 00:34:59,759
Speaker 3: like the numbers in it, so that it matches what

736
00:34:59,800 --> 00:35:02,640
Speaker 3: you see coming, for example, out of the real sun.

737
00:35:03,400 --> 00:35:06,239
Speaker 3: Then the idea is that maybe what do you think

738
00:35:06,280 --> 00:35:08,480
Speaker 3: is going on inside the sun is actually what is

739
00:35:08,520 --> 00:35:09,640
Speaker 3: going on inside the sun.

740
00:35:09,840 --> 00:35:12,600
Speaker 1: Yeah, that's exactly right. And somebody else might come up

741
00:35:12,600 --> 00:35:16,080
Speaker 1: with another simulation saying, actually, I think something else is happening.

742
00:35:16,440 --> 00:35:18,280
Speaker 1: And then you can ask, well, what's the difference between

743
00:35:18,320 --> 00:35:21,200
Speaker 1: these two simulations. Do they predict any different things that

744
00:35:21,239 --> 00:35:24,120
Speaker 1: we actually can measure? That you can go off and

745
00:35:24,239 --> 00:35:27,279
Speaker 1: use that to distinguish between two various ideas, And we

746
00:35:27,360 --> 00:35:29,640
Speaker 1: talk about this all the time on the podcast. Sometimes

747
00:35:29,680 --> 00:35:32,600
Speaker 1: we have like two different possible ideas for what's happening

748
00:35:32,719 --> 00:35:35,840
Speaker 1: near black holes. Remember we once talked about the magnetic

749
00:35:35,880 --> 00:35:38,760
Speaker 1: fields near black holes, something we could definitely not measure,

750
00:35:39,040 --> 00:35:41,399
Speaker 1: and there were two different models. One was called mad,

751
00:35:41,520 --> 00:35:44,600
Speaker 1: one was called sane, and they made slightly different predictions.

752
00:35:44,640 --> 00:35:47,239
Speaker 1: And then the recent picture of the black hole helped

753
00:35:47,320 --> 00:35:51,040
Speaker 1: us distinguish between these two models, these two simulations for

754
00:35:51,160 --> 00:35:52,600
Speaker 1: black hole magnetic fields.

755
00:35:52,880 --> 00:35:55,040
Speaker 3: Right, But I guess that's the tricky thing is like

756
00:35:55,280 --> 00:35:57,200
Speaker 3: just because the simulation matches what do you see at

757
00:35:57,239 --> 00:35:59,640
Speaker 3: the end. It may not necessarily be what's going on.

758
00:36:00,520 --> 00:36:03,400
Speaker 3: It could just be sort of a coincidence that it matches.

759
00:36:03,520 --> 00:36:05,440
Speaker 1: It certainly could be, And you always have to be

760
00:36:05,480 --> 00:36:08,800
Speaker 1: careful trusting your simulation. You always need ways to validate

761
00:36:08,800 --> 00:36:10,800
Speaker 1: it and to ensure that the bits that are important

762
00:36:10,840 --> 00:36:12,320
Speaker 1: to your science question are.

763
00:36:12,200 --> 00:36:15,480
Speaker 3: Accurate, because I guess sometimes that's the only option that

764
00:36:15,560 --> 00:36:18,359
Speaker 3: we have, right, Like you're saying, you can't just stick

765
00:36:18,400 --> 00:36:20,400
Speaker 3: a stick inside the science and see what's going on

766
00:36:21,000 --> 00:36:24,080
Speaker 3: and things like maybe black holes or the Big Bang,

767
00:36:24,120 --> 00:36:25,880
Speaker 3: Like there's no way where we can go back in

768
00:36:25,920 --> 00:36:28,120
Speaker 3: time and do an experiment on the Big Bang, right,

769
00:36:28,160 --> 00:36:30,799
Speaker 3: So we sort of have to rely on these simulations

770
00:36:30,800 --> 00:36:32,960
Speaker 3: to try to understand what was going on.

771
00:36:33,160 --> 00:36:35,760
Speaker 1: Yeah, And they've turned out to be extraordinarily powerful tools

772
00:36:35,760 --> 00:36:37,640
Speaker 1: that give us insight into what might have happened in

773
00:36:37,640 --> 00:36:39,840
Speaker 1: the early universe or what's going on in the hearts

774
00:36:39,840 --> 00:36:42,840
Speaker 1: of black holes or neutron stars. I can't really imagine

775
00:36:42,880 --> 00:36:43,839
Speaker 1: doing science without them.

776
00:36:44,040 --> 00:36:46,279
Speaker 3: All Right, Well, that's sort of what I guess a

777
00:36:46,320 --> 00:36:50,520
Speaker 3: pretty good answer for why scientists use simulations and surprise twist,

778
00:36:50,560 --> 00:36:54,280
Speaker 3: this whole conversation was just a simulation of our discussion

779
00:36:54,320 --> 00:36:56,759
Speaker 3: of the puppet. This is like a sixth sense. I

780
00:36:56,800 --> 00:36:59,759
Speaker 3: only see simulated people. This was not the real podcast, right.

781
00:37:00,040 --> 00:37:03,200
Speaker 1: That's right. Yeah, and hopefully this answer is also true

782
00:37:03,239 --> 00:37:04,160
Speaker 1: in the real universe.

783
00:37:04,280 --> 00:37:07,640
Speaker 3: But Daniel, you got to interview a scientist who does

784
00:37:07,680 --> 00:37:11,600
Speaker 3: physics and actually also wrote a book about simulating things

785
00:37:11,640 --> 00:37:12,320
Speaker 3: in the universe.

786
00:37:12,600 --> 00:37:15,640
Speaker 1: That's right. I had a fun chat with Professor Andrew Pnsen.

787
00:37:16,000 --> 00:37:19,440
Speaker 1: He's a cosmologist and a professor at University of College London,

788
00:37:19,480 --> 00:37:22,080
Speaker 1: and he wrote a new fun book called Universe in

789
00:37:22,120 --> 00:37:26,120
Speaker 1: a Box, which explores the role of simulation in cosmology

790
00:37:26,200 --> 00:37:28,840
Speaker 1: and in science in general. And he does have an

791
00:37:28,880 --> 00:37:31,680
Speaker 1: answer for the question is our universe a simulation?

792
00:37:32,040 --> 00:37:34,120
Speaker 3: Now, if I order Universe in a box? Do I

793
00:37:34,160 --> 00:37:35,360
Speaker 3: get a universe in a box?

794
00:37:37,600 --> 00:37:39,239
Speaker 1: Maybe you get a recipe for how to put a

795
00:37:39,320 --> 00:37:40,320
Speaker 1: universe into a box.

796
00:37:40,760 --> 00:37:44,040
Speaker 3: That's that's not the universe in a box? And also

797
00:37:44,080 --> 00:37:47,880
Speaker 3: why a box? Why not? I don't know a spherical container.

798
00:37:47,960 --> 00:37:49,760
Speaker 1: I thought you were going to ask, if the universe

799
00:37:49,800 --> 00:37:52,200
Speaker 1: is in a box, what universe is the box in?

800
00:37:52,680 --> 00:37:55,120
Speaker 3: Hmmm, No, I wasn't going to ask that.

801
00:37:57,400 --> 00:37:59,560
Speaker 1: Dang it, Well, my simulated Jorge wind Awry.

802
00:37:59,400 --> 00:38:02,640
Speaker 3: In the multiverse. I don't know. Aren't they like meta

803
00:38:02,719 --> 00:38:05,560
Speaker 3: universes outside of our universe? Isn't that the idea?

804
00:38:05,719 --> 00:38:08,400
Speaker 1: Click on the multiverse box option on Amazon Shipping.

805
00:38:08,480 --> 00:38:11,120
Speaker 3: The question then is can you have a multiverse in

806
00:38:11,120 --> 00:38:14,560
Speaker 3: a box? Anyways, you had a great conversation with Andrew.

807
00:38:14,760 --> 00:38:15,480
Speaker 1: I certainly did.

808
00:38:15,600 --> 00:38:17,160
Speaker 3: What motivated him to write this book.

809
00:38:17,200 --> 00:38:19,520
Speaker 1: You felt like the role of simulation in science was

810
00:38:19,640 --> 00:38:23,200
Speaker 1: super important and yet hadn't really been explored by any

811
00:38:23,200 --> 00:38:25,400
Speaker 1: pop side book, and so he wanted to share his

812
00:38:25,520 --> 00:38:27,840
Speaker 1: love for simulations with everybody.

813
00:38:27,960 --> 00:38:31,200
Speaker 3: Cool. Well, here is Daniel's interview with Professor Andrew Ponson

814
00:38:31,360 --> 00:38:34,040
Speaker 3: about his new book, Universe in a Box.

815
00:38:36,400 --> 00:38:39,360
Speaker 1: So, then it's my pleasure to welcome to the program.

816
00:38:39,440 --> 00:38:44,480
Speaker 1: Andrew Ponson, a cosmologist and professor at University College London.

817
00:38:44,600 --> 00:38:46,879
Speaker 1: And you thank you very much for joining us today. Oh,

818
00:38:46,880 --> 00:38:50,640
Speaker 1: thank you. So your book is called Universe in a Box.

819
00:38:50,760 --> 00:38:54,000
Speaker 1: It's a fascinating and compelling history and sort of definition

820
00:38:54,160 --> 00:38:57,560
Speaker 1: of what is a simulation and why it's important in science.

821
00:38:57,920 --> 00:39:01,000
Speaker 1: So let's start off with a very very bass What

822
00:39:01,360 --> 00:39:03,520
Speaker 1: is a simulation in your point of view.

823
00:39:03,640 --> 00:39:05,600
Speaker 9: There are different definitions you can give, but I think

824
00:39:05,640 --> 00:39:08,400
Speaker 9: a good place to start is by thinking it's trying

825
00:39:08,440 --> 00:39:13,520
Speaker 9: to capture some element of the real world inside a computer,

826
00:39:13,760 --> 00:39:16,560
Speaker 9: and that can take many different forms. It doesn't even

827
00:39:16,640 --> 00:39:19,400
Speaker 9: have to be a physics right, it's we can have

828
00:39:19,440 --> 00:39:23,600
Speaker 9: simulations of something like human behavior. There are simulations of

829
00:39:23,600 --> 00:39:26,759
Speaker 9: the way that crowds might behave that architects use to

830
00:39:27,040 --> 00:39:30,279
Speaker 9: make safer buildings by making the passageways the right kind

831
00:39:30,280 --> 00:39:33,320
Speaker 9: of size and shape that if there's an emergency situation,

832
00:39:33,560 --> 00:39:37,280
Speaker 9: then humans will evacuate the building in the most efficient,

833
00:39:37,360 --> 00:39:40,400
Speaker 9: safe way. But I think what that already teaches you

834
00:39:40,560 --> 00:39:42,920
Speaker 9: is that it's possible to do a simulation of something

835
00:39:43,480 --> 00:39:48,000
Speaker 9: without necessarily understanding everything about that thing before you start.

836
00:39:48,760 --> 00:39:52,040
Speaker 9: Because if you think of crowds, they're made out of people.

837
00:39:52,640 --> 00:39:56,840
Speaker 9: We can't actually predict everything about how an individual human's

838
00:39:56,880 --> 00:40:00,200
Speaker 9: going to react in any given scenario. And yet yet

839
00:40:00,480 --> 00:40:03,440
Speaker 9: we can make simulations that are useful. They might not

840
00:40:03,480 --> 00:40:06,680
Speaker 9: be perfect, but they're useful for giving us some insight

841
00:40:06,760 --> 00:40:09,960
Speaker 9: into the way that crowds might behave. So when you

842
00:40:10,000 --> 00:40:13,080
Speaker 9: take that across to the physics environment, and in particular

843
00:40:13,160 --> 00:40:18,480
Speaker 9: my area, cosmology, it's about trying to capture something about

844
00:40:18,520 --> 00:40:21,920
Speaker 9: how the universe behaves. But we know from the outset

845
00:40:21,960 --> 00:40:23,400
Speaker 9: we're never going to get that perfect.

846
00:40:23,640 --> 00:40:26,719
Speaker 1: So then where is the value in a simulation If

847
00:40:26,760 --> 00:40:29,840
Speaker 1: you have to, like encode in already what's going to

848
00:40:29,880 --> 00:40:32,840
Speaker 1: happen when people bump into each other, or the purchasing

849
00:40:32,960 --> 00:40:36,319
Speaker 1: choices of people, or how galaxies interact. If you have

850
00:40:36,320 --> 00:40:38,920
Speaker 1: to already build in the physics, what are you learning

851
00:40:38,960 --> 00:40:41,680
Speaker 1: from the simulation? How do you get any information out

852
00:40:41,680 --> 00:40:41,960
Speaker 1: of it?

853
00:40:42,680 --> 00:40:45,799
Speaker 9: Well, the point is that you code in some things

854
00:40:45,840 --> 00:40:49,719
Speaker 9: about how the individual bits within your simulation behave. I

855
00:40:49,719 --> 00:40:51,480
Speaker 9: guess you know, in the case of the crowd, that

856
00:40:51,520 --> 00:40:54,960
Speaker 9: would be how an individual human might behave under a

857
00:40:55,080 --> 00:40:58,280
Speaker 9: variety of circumstances. Are that in the case of physics,

858
00:40:58,280 --> 00:41:01,840
Speaker 9: it might be how we think dark matter particles flow

859
00:41:01,920 --> 00:41:04,800
Speaker 9: through the universe and interact with each other through gravity.

860
00:41:04,960 --> 00:41:08,200
Speaker 9: And what the simulation does is take a very large

861
00:41:08,320 --> 00:41:12,320
Speaker 9: number of those elements and kind of have them all

862
00:41:12,360 --> 00:41:16,319
Speaker 9: individually doing their thing. But the behavior that then emerges

863
00:41:17,080 --> 00:41:20,719
Speaker 9: can be very hard to anticipate in advance. And that's

864
00:41:20,760 --> 00:41:24,719
Speaker 9: the point. Understanding how a crowd behaves is not at

865
00:41:24,760 --> 00:41:27,280
Speaker 9: all the same thing as understanding how a human behaves,

866
00:41:27,880 --> 00:41:31,680
Speaker 9: and in the same way, understanding how dark matter behaves

867
00:41:31,800 --> 00:41:35,000
Speaker 9: through our whole cosmos is not at all the same

868
00:41:35,040 --> 00:41:39,040
Speaker 9: thing as understanding what an individual particle of dark matter

869
00:41:39,120 --> 00:41:39,560
Speaker 9: might do.

870
00:41:39,840 --> 00:41:43,399
Speaker 1: This principle of emergent phenomena is something I'm super fascinated by.

871
00:41:43,840 --> 00:41:46,319
Speaker 1: It's incredible to me that sometimes we have laws of

872
00:41:46,320 --> 00:41:50,640
Speaker 1: physics at one scale, which you know, causally determine different

873
00:41:50,680 --> 00:41:53,799
Speaker 1: sort of laws of physics at another scale, which we

874
00:41:53,840 --> 00:41:56,040
Speaker 1: can't always easily predict. But as you say, we can

875
00:41:56,200 --> 00:41:59,480
Speaker 1: observe in action if we can, you know, construct The

876
00:41:59,560 --> 00:42:02,759
Speaker 1: right set to you is simulation. Is it a kind

877
00:42:02,800 --> 00:42:05,440
Speaker 1: of experiment, Is it a kind of theory or is

878
00:42:05,480 --> 00:42:07,640
Speaker 1: it sort of a new branch of science.

879
00:42:08,000 --> 00:42:09,759
Speaker 9: I think it's a new branch, but I think it's

880
00:42:09,760 --> 00:42:12,719
Speaker 9: got something in common with theory and with experiments, and

881
00:42:12,760 --> 00:42:15,719
Speaker 9: I guess I tilt mainly towards thinking of it as

882
00:42:15,719 --> 00:42:20,160
Speaker 9: an experiment. Now that's a little bit controversial. Sometimes people say, well,

883
00:42:20,160 --> 00:42:23,120
Speaker 9: it can't be an experiment. You've told the computer what

884
00:42:23,239 --> 00:42:26,600
Speaker 9: to do, whereas in an experiment you're supposed to go

885
00:42:26,640 --> 00:42:29,520
Speaker 9: and ask nature. You know, you're supposed to put things

886
00:42:29,560 --> 00:42:31,880
Speaker 9: to the test and confront them with the reality of

887
00:42:31,920 --> 00:42:35,839
Speaker 9: how things really work. But you know, I'm not sure

888
00:42:35,840 --> 00:42:38,920
Speaker 9: that that distinction is always so clear. So an example

889
00:42:38,960 --> 00:42:41,719
Speaker 9: that I give in the book is, let's say you're

890
00:42:42,040 --> 00:42:45,759
Speaker 9: just trying to build an aircraft and you have some

891
00:42:45,880 --> 00:42:48,400
Speaker 9: idea about how you want to shape the wing, but

892
00:42:48,440 --> 00:42:51,080
Speaker 9: you don't know exactly how that wing is going to perform.

893
00:42:51,400 --> 00:42:54,759
Speaker 9: Now you now have a choice. You could build a

894
00:42:54,800 --> 00:42:57,640
Speaker 9: scale model of your wing and put it inside a

895
00:42:57,680 --> 00:43:01,280
Speaker 9: wind tunnel and see how it performs inside a wind tunnel,

896
00:43:01,280 --> 00:43:04,560
Speaker 9: and that's kind of an experiment. Or you could make

897
00:43:04,560 --> 00:43:07,920
Speaker 9: a digital version of your wing and put it inside

898
00:43:07,960 --> 00:43:11,799
Speaker 9: an airflow inside a computer there's a kind of simulated

899
00:43:11,840 --> 00:43:15,040
Speaker 9: airflow and see how it behaves there. And both of

900
00:43:15,080 --> 00:43:18,160
Speaker 9: those are going to have limitations. There's definitely limitations on

901
00:43:18,200 --> 00:43:20,560
Speaker 9: what you can achieve inside the computer, but there's also

902
00:43:20,640 --> 00:43:24,440
Speaker 9: limitations in what you can achieve in a wind tunnel.

903
00:43:24,560 --> 00:43:27,600
Speaker 9: You can't make an infinitely big wind tunnel. It's going

904
00:43:27,640 --> 00:43:30,360
Speaker 9: to have edges, things are going to be the wrong scale.

905
00:43:31,200 --> 00:43:34,919
Speaker 9: So when you do experiments, you are making some set

906
00:43:34,920 --> 00:43:39,080
Speaker 9: of assumptions about the real world and how what you're

907
00:43:39,120 --> 00:43:41,879
Speaker 9: doing applies to the real world, and I think that's

908
00:43:41,960 --> 00:43:44,600
Speaker 9: just that's true in simulations as well. So overall, this

909
00:43:44,719 --> 00:43:46,719
Speaker 9: is why I start to think more and more of

910
00:43:46,760 --> 00:43:49,360
Speaker 9: simulations as types of experiment.

911
00:43:49,560 --> 00:43:52,480
Speaker 1: I have an argument with my brother who's a computer scientist,

912
00:43:52,800 --> 00:43:56,080
Speaker 1: and he runs what he calls experiments on his machine

913
00:43:56,160 --> 00:43:59,440
Speaker 1: learning models. I'm like, that's not an experiment. You're just

914
00:43:59,520 --> 00:44:02,600
Speaker 1: doing it in your computer. But you're absolutely right that

915
00:44:02,719 --> 00:44:04,480
Speaker 1: if you don't know the outcome and you're learning something,

916
00:44:04,520 --> 00:44:07,160
Speaker 1: it can be considered also an experiment. But you mentioned

917
00:44:07,200 --> 00:44:08,960
Speaker 1: something which I wanted to ask you about anyway, which

918
00:44:08,960 --> 00:44:13,160
Speaker 1: is the limitation of simulation. You're concocting sort of an

919
00:44:13,280 --> 00:44:17,239
Speaker 1: artificial universe, and you're learning something about that universe. If

920
00:44:17,239 --> 00:44:20,359
Speaker 1: that universe doesn't follow the same rules as ours, then

921
00:44:20,400 --> 00:44:23,759
Speaker 1: obviously we're not learning something about reality, which usually is

922
00:44:23,840 --> 00:44:26,719
Speaker 1: the goal. Can you say something about how we know

923
00:44:27,080 --> 00:44:31,680
Speaker 1: when to trust our simulations with the fundamental limitations of simulation.

924
00:44:31,320 --> 00:44:34,640
Speaker 9: Are Yeah, I mean that is the hardest, most difficult question.

925
00:44:34,680 --> 00:44:37,719
Speaker 9: At the heart of doing good simulation is knowing what

926
00:44:37,760 --> 00:44:40,799
Speaker 9: to trust and what not to trust. And often it's hard,

927
00:44:40,960 --> 00:44:44,319
Speaker 9: you know, it can be really hard to know. I mean, fundamentally,

928
00:44:44,360 --> 00:44:49,200
Speaker 9: the limitation is just computational power that even if you're

929
00:44:49,239 --> 00:44:51,640
Speaker 9: doing something like a weather forecast, which I talk about

930
00:44:51,680 --> 00:44:54,919
Speaker 9: a bit in the book. You know, just the atmosphere

931
00:44:55,000 --> 00:44:58,279
Speaker 9: of the Earth has so many molecules in it that

932
00:44:58,360 --> 00:45:01,080
Speaker 9: you're never going to track each ind vidual molecule, right,

933
00:45:01,160 --> 00:45:04,080
Speaker 9: So you're going to have to make some kind of approximation.

934
00:45:04,120 --> 00:45:06,840
Speaker 9: You're going to parcel up the air into almost like

935
00:45:06,920 --> 00:45:10,600
Speaker 9: big hypothetical bags of air that move around through our

936
00:45:10,640 --> 00:45:15,319
Speaker 9: atmosphere and use some laws to describe that. But then

937
00:45:15,320 --> 00:45:17,680
Speaker 9: you're going to have to go and say, well, you know,

938
00:45:17,719 --> 00:45:20,239
Speaker 9: we're not getting all the small scale details, right, We're

939
00:45:20,239 --> 00:45:23,000
Speaker 9: going to have to put in some corrections. In the

940
00:45:23,000 --> 00:45:27,360
Speaker 9: case of meteorology, even clouds can be quite hard to

941
00:45:27,440 --> 00:45:30,120
Speaker 9: predict because you're just not getting all of those tiny

942
00:45:30,200 --> 00:45:33,880
Speaker 9: details that contribute to the way that a cloud forms

943
00:45:33,920 --> 00:45:36,640
Speaker 9: in reality. So you need to go in and put

944
00:45:36,840 --> 00:45:39,920
Speaker 9: into your simulation some kind of correction almost by hand.

945
00:45:40,520 --> 00:45:44,200
Speaker 9: You say, under these circumstances, clouds must start to form.

946
00:45:44,960 --> 00:45:48,440
Speaker 9: And you know, if you're a weather forecaster, you see,

947
00:45:48,840 --> 00:45:52,000
Speaker 9: how well did I do by making that assumption about

948
00:45:52,000 --> 00:45:55,279
Speaker 9: how clouds form, and over time you sort of incrementally

949
00:45:55,320 --> 00:45:59,640
Speaker 9: improve by comparing how your simulation did with the reality

950
00:45:59,680 --> 00:46:02,760
Speaker 9: of how the weather unfolded. So we can do something

951
00:46:02,920 --> 00:46:05,439
Speaker 9: a bit similar in cosmology. It's not quite the same

952
00:46:05,520 --> 00:46:08,000
Speaker 9: because we don't get to kind of do the repeat

953
00:46:08,080 --> 00:46:10,239
Speaker 9: experiments in quite the same way as you do in

954
00:46:10,560 --> 00:46:13,560
Speaker 9: weather forecasting, for example. In some level, it's the same

955
00:46:13,680 --> 00:46:17,040
Speaker 9: kind of iterative process that we're getting better over time.

956
00:46:17,120 --> 00:46:19,960
Speaker 9: We're understanding the way that we have to put in

957
00:46:20,000 --> 00:46:24,359
Speaker 9: corrections to our simulations to account for things like the

958
00:46:24,400 --> 00:46:27,760
Speaker 9: way that stars evolve and change over time and dump

959
00:46:27,920 --> 00:46:30,920
Speaker 9: energy into the universe, and what black holes are up to,

960
00:46:31,000 --> 00:46:33,640
Speaker 9: and all of these things that we actually have to

961
00:46:33,880 --> 00:46:35,719
Speaker 9: help the computer along the way, if you like.

962
00:46:35,920 --> 00:46:38,399
Speaker 1: So why is it that we need to make these

963
00:46:38,440 --> 00:46:42,880
Speaker 1: corrections we have these limitations. Is it purely just computational power.

964
00:46:43,239 --> 00:46:47,120
Speaker 1: In the limit of infinite computing power, could we predict

965
00:46:47,120 --> 00:46:50,479
Speaker 1: the weather tomorrow starting from particle physics and modeling every

966
00:46:50,480 --> 00:46:53,759
Speaker 1: single quirk in the atmosphere, Or is there a conceptual

967
00:46:53,800 --> 00:46:56,640
Speaker 1: limit there some obstacle that we can't overcome even with

968
00:46:56,680 --> 00:46:57,480
Speaker 1: infinite computing.

969
00:46:58,040 --> 00:47:00,640
Speaker 9: I think both are a problem. So, first of all,

970
00:47:00,719 --> 00:47:03,719
Speaker 9: we are very far from having infinite computer power, a

971
00:47:03,880 --> 00:47:06,479
Speaker 9: very very long way away from that. But secondly, you're right,

972
00:47:06,560 --> 00:47:10,160
Speaker 9: I mean there are more fundamental limitations as well. In particular,

973
00:47:10,320 --> 00:47:13,840
Speaker 9: we do not know exactly where every molecule is in

974
00:47:13,880 --> 00:47:17,920
Speaker 9: the atmosphere to start with. So even if you had

975
00:47:17,920 --> 00:47:21,680
Speaker 9: a computer powerful enough to track at the molecular level

976
00:47:21,719 --> 00:47:24,279
Speaker 9: what our atmosphere is doing, you wouldn't know how to

977
00:47:24,800 --> 00:47:28,439
Speaker 9: start the simulation. You wouldn't have enough data to tell

978
00:47:28,480 --> 00:47:31,640
Speaker 9: it what the atmosphere looks like today. So there's an

979
00:47:31,680 --> 00:47:35,239
Speaker 9: inaccuracy that's sort of just coming from not having that

980
00:47:35,360 --> 00:47:39,160
Speaker 9: perfect data, and an effect known as chaos means that

981
00:47:39,840 --> 00:47:45,000
Speaker 9: imperfect initial data very quickly turns into big errors. So

982
00:47:45,320 --> 00:47:48,360
Speaker 9: there's a kind of famous example of this. It was

983
00:47:48,520 --> 00:47:51,839
Speaker 9: Edward Lorenz who sort of gave the thought experiment of

984
00:47:51,880 --> 00:47:56,480
Speaker 9: a butterfly flapping its wings somewhere in Europe, say, and

985
00:47:56,560 --> 00:47:59,040
Speaker 9: it has a sort of series of knock on effects

986
00:47:59,040 --> 00:48:02,160
Speaker 9: that over time just amplify and amplify, and eventually the

987
00:48:02,360 --> 00:48:07,759
Speaker 9: tiny little gust from the butterfly's wings actually stimulates the

988
00:48:07,800 --> 00:48:12,040
Speaker 9: formation of a hurricane. And you know, these kind of effects,

989
00:48:12,120 --> 00:48:14,880
Speaker 9: we know they're there and physics. We call them chaos,

990
00:48:15,320 --> 00:48:20,200
Speaker 9: and so you know, the slightest inaccuracy in how you

991
00:48:20,239 --> 00:48:25,680
Speaker 9: set things up will eventually make the simulation depart from reality.

992
00:48:25,960 --> 00:48:28,520
Speaker 9: So we know that's true in cosmology as well. And

993
00:48:28,640 --> 00:48:33,040
Speaker 9: we don't have, you know, perfect information about the early universe.

994
00:48:33,120 --> 00:48:36,040
Speaker 9: We have quite good ideas for what was going on there,

995
00:48:36,560 --> 00:48:40,880
Speaker 9: but it's imperfect, and that means because of chaos and

996
00:48:40,920 --> 00:48:44,680
Speaker 9: the way that those imperfections are amplified over time, what

997
00:48:44,719 --> 00:48:47,480
Speaker 9: we end up with is, in some sense, it's like

998
00:48:47,480 --> 00:48:52,560
Speaker 9: a statistical recreation of the universe. It's telling us statistically

999
00:48:53,320 --> 00:48:56,080
Speaker 9: what sorts of things should be in the universe, in

1000
00:48:56,120 --> 00:48:59,680
Speaker 9: what sorts of mixtures, in what kind of patterns, rather

1001
00:48:59,719 --> 00:49:03,279
Speaker 9: than literally recreating the universe. So it's almost more like

1002
00:49:03,360 --> 00:49:08,520
Speaker 9: sort of climate it's like a climate simulation almost, rather

1003
00:49:08,560 --> 00:49:09,640
Speaker 9: than a weather simulation.

1004
00:49:10,200 --> 00:49:12,759
Speaker 1: I'm very interested in the history of simulations as well.

1005
00:49:12,800 --> 00:49:16,080
Speaker 1: I mean, theoretical science is like thousands of years old.

1006
00:49:16,160 --> 00:49:19,240
Speaker 1: Experimental science people argue about might be hundreds of years old.

1007
00:49:19,400 --> 00:49:22,799
Speaker 1: Simulation based science seems like decades old. Can you take

1008
00:49:22,880 --> 00:49:25,280
Speaker 1: us back to the root of it, Where does it begin? Really?

1009
00:49:25,400 --> 00:49:27,680
Speaker 9: Well, I think you know the very earliest route you

1010
00:49:27,719 --> 00:49:31,160
Speaker 9: can find is in the nineteenth century, where Charles Babbage

1011
00:49:31,200 --> 00:49:33,920
Speaker 9: and Ada Lovelace were working on the idea of a

1012
00:49:33,960 --> 00:49:37,480
Speaker 9: computer very similar to our modern computers. It was the

1013
00:49:37,520 --> 00:49:40,480
Speaker 9: first time, really anybody expressed the idea of having a

1014
00:49:40,560 --> 00:49:46,279
Speaker 9: machine that could be told to perform any calculation. So

1015
00:49:46,400 --> 00:49:49,440
Speaker 9: before that there were machines that did specific calculations, but

1016
00:49:50,000 --> 00:49:53,040
Speaker 9: this was the first time somebody envisaged a machine where

1017
00:49:53,160 --> 00:49:55,400
Speaker 9: you could just give it instructions and it would carry

1018
00:49:55,400 --> 00:50:00,279
Speaker 9: out calculations to your specifications. And Ada Lovelace actually he

1019
00:50:00,320 --> 00:50:03,480
Speaker 9: wrote at that time that one of the applications of

1020
00:50:03,520 --> 00:50:06,440
Speaker 9: being able to do that was to be able to

1021
00:50:06,480 --> 00:50:10,719
Speaker 9: take what we think are the governing laws of physics

1022
00:50:11,400 --> 00:50:13,960
Speaker 9: and make them kind of practical. You know, get the

1023
00:50:14,040 --> 00:50:18,080
Speaker 9: computer to do all of the calculations that turns those

1024
00:50:18,440 --> 00:50:25,160
Speaker 9: abstract equations into concrete, specific predictions for different scenarios. So

1025
00:50:25,239 --> 00:50:29,000
Speaker 9: that's probably the first time anyone expressed what we would

1026
00:50:29,000 --> 00:50:32,800
Speaker 9: recognize as a modern simulation. Then, in terms of actually

1027
00:50:33,239 --> 00:50:37,239
Speaker 9: performing simulations, remarkably, some people tried to do this in

1028
00:50:37,239 --> 00:50:43,640
Speaker 9: the twentieth century before digital computers were actually made. So

1029
00:50:43,840 --> 00:50:49,000
Speaker 9: there are some beautiful stories like a crazy character called

1030
00:50:49,080 --> 00:50:52,680
Speaker 9: Lewis Fry Richardson who was actually on the front line

1031
00:50:52,719 --> 00:50:59,080
Speaker 9: of World War One trying to calculate weather forecasts using

1032
00:50:59,200 --> 00:51:02,960
Speaker 9: pen and paper, very very repetitive calculations. He was doing

1033
00:51:03,520 --> 00:51:05,960
Speaker 9: that would be exactly what a computer does today to

1034
00:51:05,960 --> 00:51:08,759
Speaker 9: do a weather forecast, but he was doing it just

1035
00:51:08,920 --> 00:51:11,880
Speaker 9: with pen and paper and taking him, you know, weeks,

1036
00:51:11,920 --> 00:51:15,759
Speaker 9: stretching out into years just to do one forecast. He

1037
00:51:15,800 --> 00:51:17,360
Speaker 9: wasn't trying to be practical about it. He was just

1038
00:51:17,400 --> 00:51:20,800
Speaker 9: trying to prove a point that this is actually doable

1039
00:51:20,840 --> 00:51:23,960
Speaker 9: in practice. And then you know, by the end of

1040
00:51:23,960 --> 00:51:28,120
Speaker 9: World War Two there were actual computers available, and very

1041
00:51:28,160 --> 00:51:31,759
Speaker 9: quickly from there the whole business of simulating all sorts

1042
00:51:31,760 --> 00:51:35,200
Speaker 9: of different things, but ultimately the entire universe kind of

1043
00:51:35,200 --> 00:51:36,560
Speaker 9: grew up quite quickly from there.

1044
00:51:40,920 --> 00:51:42,719
Speaker 1: When you pop a piece of cheese into your mouth,

1045
00:51:42,840 --> 00:51:45,960
Speaker 1: or enjoy a rich spoonful of greeky yogurt, you're probably

1046
00:51:46,000 --> 00:51:50,040
Speaker 1: not thinking about the environmental impact of each and every bite,

1047
00:51:50,080 --> 00:51:52,720
Speaker 1: But the people in the dairy industry are. US Dairy

1048
00:51:52,760 --> 00:51:57,040
Speaker 1: has set themselves some ambitious sustainability goals, including being greenhouse

1049
00:51:57,080 --> 00:51:59,640
Speaker 1: gas neutral by twenty to fifty That's why they're working

1050
00:51:59,640 --> 00:52:02,000
Speaker 1: hard every day to find new ways to reduce waste,

1051
00:52:02,080 --> 00:52:06,279
Speaker 1: conserve natural resources, and drive down greenhouse gas emissions. Take water,

1052
00:52:06,360 --> 00:52:09,440
Speaker 1: for example, most dairy farms reuse water up to four

1053
00:52:09,480 --> 00:52:12,960
Speaker 1: times the same water cools the milk, cleans equipment, washes

1054
00:52:13,000 --> 00:52:15,800
Speaker 1: the barn, and irrigates the crops. How is US dairy

1055
00:52:15,840 --> 00:52:19,560
Speaker 1: tackling greenhouse gases. Many farms use anaerobic digestors that turn

1056
00:52:19,640 --> 00:52:23,520
Speaker 1: the methane from maneuver into renewable energy that can power farms, towns,

1057
00:52:23,560 --> 00:52:25,640
Speaker 1: and electric cars. So the next time you grab a

1058
00:52:25,640 --> 00:52:27,680
Speaker 1: slice of pizza or lick an ice cream cone, know

1059
00:52:27,760 --> 00:52:30,439
Speaker 1: that dairy farmers and processors around the country are using

1060
00:52:30,480 --> 00:52:33,959
Speaker 1: the latest practices and innovations to provide the nutrient dense

1061
00:52:34,080 --> 00:52:36,799
Speaker 1: dairy products we love with less of an impact. Visit

1062
00:52:36,880 --> 00:52:39,680
Speaker 1: us dairy dot com slash sustainability to learn more.

1063
00:52:40,719 --> 00:52:44,240
Speaker 10: There are children, friends, and families walking, riding on passing

1064
00:52:44,239 --> 00:52:46,680
Speaker 10: the roads every day. Remember they're real people with loved

1065
00:52:46,680 --> 00:52:48,880
Speaker 10: ones who need them to get home safely. Protect our

1066
00:52:48,920 --> 00:52:51,000
Speaker 10: cyclists and pedestrians because they're people too.

1067
00:52:51,280 --> 00:52:51,800
Speaker 1: Go safely.

1068
00:52:51,880 --> 00:52:55,279
Speaker 10: California From the California Office of Traffic Safety and caltrans.

1069
00:52:54,840 --> 00:52:57,640
Speaker 11: Time is a luxury for us, especially if you're a mom.

1070
00:52:57,880 --> 00:53:00,760
Speaker 11: That's why we need a skincare routine that's easy, fast,

1071
00:53:00,920 --> 00:53:04,040
Speaker 11: and gives us results. Plus, what if your products had

1072
00:53:04,120 --> 00:53:07,920
Speaker 11: thousands of five star reviews, we're natural and affordable, Well

1073
00:53:08,120 --> 00:53:11,799
Speaker 11: say hello to dime Beauty. Dime Beauty is clean, high

1074
00:53:11,880 --> 00:53:15,040
Speaker 11: end skincare that is affordable and it really works. Not

1075
00:53:15,080 --> 00:53:17,960
Speaker 11: sure where to start? I highly recommend the work system.

1076
00:53:18,200 --> 00:53:21,520
Speaker 11: It's everything you need in one powerful package. Take out

1077
00:53:21,520 --> 00:53:24,280
Speaker 11: the guestwork with a proven routine that includes a gentle

1078
00:53:24,360 --> 00:53:28,400
Speaker 11: yet effective cleanser, a super skin toner, two incredible serums,

1079
00:53:28,520 --> 00:53:32,480
Speaker 11: and two luxurious moisturizers. See what everyone is raving about.

1080
00:53:32,640 --> 00:53:35,759
Speaker 11: From serum sets to the always sold out retinol alternative

1081
00:53:35,760 --> 00:53:39,480
Speaker 11: TBT cream, you'll find your perfect skincare match. Dime has

1082
00:53:39,520 --> 00:53:42,680
Speaker 11: over two million happy customers and their product reviews are

1083
00:53:42,719 --> 00:53:46,200
Speaker 11: literally five stars. Love your skin again. Go to din

1084
00:53:46,280 --> 00:53:50,080
Speaker 11: beautyco dot com for twenty percent off with code get Dime.

1085
00:53:50,360 --> 00:53:54,120
Speaker 11: That's din beautyco dot com code get Dime for twenty

1086
00:53:54,120 --> 00:53:54,760
Speaker 11: percent off.

1087
00:53:54,960 --> 00:53:58,759
Speaker 12: It's time for today's Lucky Land horoscope with Victoria Cash.

1088
00:53:59,040 --> 00:54:01,880
Speaker 12: Life's gotten mundane, So shake up the daily routine and

1089
00:54:01,920 --> 00:54:04,560
Speaker 12: be adventurous with a trip to Lucky Land. You know

1090
00:54:04,640 --> 00:54:07,560
Speaker 12: what they say, Your chance to win starts with a spin,

1091
00:54:07,840 --> 00:54:10,200
Speaker 12: So go to Lucky landslots dot com to play over

1092
00:54:10,239 --> 00:54:12,799
Speaker 12: one hundred social casino style games for free for your

1093
00:54:12,880 --> 00:54:18,120
Speaker 12: chance to redeem some serious prizes. Get lucky today at

1094
00:54:18,200 --> 00:54:22,000
Speaker 12: Lucky landslots dot com. No purchase necessary. BGW group void

1095
00:54:22,000 --> 00:54:24,439
Speaker 12: were prohibited by Law eighteen plus terms of conditions apply.

1096
00:54:33,600 --> 00:54:36,160
Speaker 1: Tell us more about the role of simulation in your

1097
00:54:36,200 --> 00:54:39,239
Speaker 1: personal research. Is this something you explore because you're fascinated

1098
00:54:39,239 --> 00:54:41,319
Speaker 1: by the computer science of it, or to use it

1099
00:54:41,440 --> 00:54:43,800
Speaker 1: just a tool that helps answer your physics questions.

1100
00:54:44,200 --> 00:54:46,279
Speaker 9: I think it depends on the day you ask me.

1101
00:54:46,320 --> 00:54:50,440
Speaker 9: I mean, some days I really enjoy the computer science

1102
00:54:50,480 --> 00:54:53,200
Speaker 9: of all of this, and you know, it's undeniably cool

1103
00:54:53,280 --> 00:54:55,400
Speaker 9: to get to work with some of the world's biggest

1104
00:54:55,800 --> 00:55:00,000
Speaker 9: supercomputers and be able to instruct them to carry out

1105
00:55:00,120 --> 00:55:04,080
Speaker 9: these kind of simulations, and the results are enormous fun

1106
00:55:04,280 --> 00:55:06,239
Speaker 9: to work with as well. So there is a bit

1107
00:55:06,239 --> 00:55:08,960
Speaker 9: of there's a bit of that kind of nerdery in it,

1108
00:55:09,040 --> 00:55:12,200
Speaker 9: But I think ultimately the thing that really keeps me

1109
00:55:12,280 --> 00:55:16,359
Speaker 9: hooked is the idea that we are contributing to a

1110
00:55:16,400 --> 00:55:20,680
Speaker 9: bigger picture of how our universe evolved, of the role

1111
00:55:20,880 --> 00:55:24,920
Speaker 9: for materials in it that we as yet don't understand.

1112
00:55:25,480 --> 00:55:29,080
Speaker 9: Things like dark matter and dark energy. That we know

1113
00:55:29,200 --> 00:55:31,240
Speaker 9: they're out there, they seem to be having a profound

1114
00:55:31,280 --> 00:55:33,880
Speaker 9: effect on our universe, but we really don't know what

1115
00:55:34,000 --> 00:55:36,640
Speaker 9: they are, and we're trying to learn more about that.

1116
00:55:37,040 --> 00:55:40,480
Speaker 9: And then I suppose ultimately what we're building towards is

1117
00:55:40,480 --> 00:55:43,160
Speaker 9: a better understanding of where we came from. You know,

1118
00:55:43,640 --> 00:55:47,360
Speaker 9: the existence of us carbon based life forms on this

1119
00:55:47,480 --> 00:55:51,120
Speaker 9: rocky planet is part of the story that we're telling.

1120
00:55:51,320 --> 00:55:54,759
Speaker 9: Because the chemical elements from which our planet and life

1121
00:55:54,800 --> 00:55:59,120
Speaker 9: are constructed weren't there in the big Bang. They've been

1122
00:55:59,160 --> 00:56:02,960
Speaker 9: manufactured over time. They need very specific conditions to be

1123
00:56:03,040 --> 00:56:08,960
Speaker 9: manufactured and then concentrated enough to start forming planets and

1124
00:56:09,120 --> 00:56:11,920
Speaker 9: enabling life and so on. So I think, you know, ultimately,

1125
00:56:12,000 --> 00:56:14,120
Speaker 9: that's the thing I'm most excited by, that we are

1126
00:56:14,239 --> 00:56:18,200
Speaker 9: telling this bigger story that in the end speaks to

1127
00:56:18,280 --> 00:56:20,920
Speaker 9: a kind of deep question within all of us about

1128
00:56:20,920 --> 00:56:21,960
Speaker 9: where did we come from?

1129
00:56:22,040 --> 00:56:26,040
Speaker 1: Have you seen yet machine learning being used to amplify

1130
00:56:26,360 --> 00:56:30,680
Speaker 1: or speed up or just overall enhance simulations in your research.

1131
00:56:31,400 --> 00:56:34,320
Speaker 9: Yeah, I mean, machine learning is more and more important

1132
00:56:34,640 --> 00:56:38,080
Speaker 9: throughout astrophysics, So it's being used in a variety of

1133
00:56:38,080 --> 00:56:42,319
Speaker 9: different ways. One is to try and improve on some

1134
00:56:42,360 --> 00:56:44,720
Speaker 9: of these things that we were talking about a moment

1135
00:56:44,760 --> 00:56:48,279
Speaker 9: ago about you know, what do you do about the

1136
00:56:48,360 --> 00:56:50,239
Speaker 9: things that you can't quite get right?

1137
00:56:50,760 --> 00:56:52,520
Speaker 3: Like the way that.

1138
00:56:52,520 --> 00:56:56,920
Speaker 9: Stars form out of gas, just such a complicated process

1139
00:56:57,040 --> 00:56:59,879
Speaker 9: that we can't capture it perfectly. The way that those

1140
00:57:00,000 --> 00:57:03,239
Speaker 9: stars then put energy back into the galaxy that they're

1141
00:57:03,239 --> 00:57:06,359
Speaker 9: forming within the roles of black holes. All of these

1142
00:57:06,360 --> 00:57:11,960
Speaker 9: things that are very complicated and multifaceted. We can use

1143
00:57:12,200 --> 00:57:15,360
Speaker 9: machine learning to do some of the hard work for

1144
00:57:15,560 --> 00:57:21,480
Speaker 9: us to learn from examples of individual simulated stars, say

1145
00:57:21,680 --> 00:57:24,880
Speaker 9: about how they behave, and kind of learn the lessons

1146
00:57:24,920 --> 00:57:27,400
Speaker 9: from those and then take them and put them in

1147
00:57:27,400 --> 00:57:30,480
Speaker 9: the bigger setting of trying to simulate then hundreds of

1148
00:57:30,520 --> 00:57:33,400
Speaker 9: billions of stars across a galaxy or maybe you know,

1149
00:57:33,480 --> 00:57:36,960
Speaker 9: even out into the universe. So machine learning can kind

1150
00:57:37,000 --> 00:57:41,160
Speaker 9: of help us take lessons from one bit of our

1151
00:57:41,280 --> 00:57:44,960
Speaker 9: simulations or one bit of physics and insert them in

1152
00:57:45,000 --> 00:57:49,480
Speaker 9: an efficient way into other simulations. That's one role for them,

1153
00:57:50,200 --> 00:57:54,080
Speaker 9: but they're also crucial in interpreting the data we get

1154
00:57:54,080 --> 00:57:57,960
Speaker 9: from the real universe, So the data that is coming

1155
00:57:58,440 --> 00:58:03,320
Speaker 9: from telescope and especially big new survey telescopes, things like

1156
00:58:03,480 --> 00:58:08,920
Speaker 9: Euclid and the Verra Rubin Observatory, these giant efforts to

1157
00:58:09,000 --> 00:58:12,400
Speaker 9: scan the sky and build maps of our universe. They

1158
00:58:12,480 --> 00:58:16,880
Speaker 9: need machine learning because the machine learning can kind of

1159
00:58:16,960 --> 00:58:20,760
Speaker 9: do a lot of the initial data processing, figuring out

1160
00:58:21,520 --> 00:58:25,880
Speaker 9: what's interesting, what we're actually looking at, where it is

1161
00:58:25,920 --> 00:58:29,720
Speaker 9: in three D space, and what needs to be flagged

1162
00:58:29,760 --> 00:58:32,800
Speaker 9: for further human follow up. All of these things, machine

1163
00:58:32,880 --> 00:58:35,600
Speaker 9: learning is playing an increasingly important role.

1164
00:58:35,880 --> 00:58:37,920
Speaker 1: And in your view, what does a future hold for

1165
00:58:38,080 --> 00:58:41,360
Speaker 1: simulation based science? Are there fields of science that don't

1166
00:58:41,440 --> 00:58:44,120
Speaker 1: yet use any simulation and are in the cusp of

1167
00:58:44,160 --> 00:58:47,160
Speaker 1: being revolutionized by this powerful new tool. I mean, I'm not.

1168
00:58:47,160 --> 00:58:50,520
Speaker 9: Aware of fields that have shied away from simulation. I

1169
00:58:50,520 --> 00:58:53,360
Speaker 9: think it is such a powerful tool that when you

1170
00:58:53,400 --> 00:58:55,880
Speaker 9: start looking for it, you do find it in use

1171
00:58:56,360 --> 00:58:59,840
Speaker 9: absolutely everywhere. But I think what we can say about

1172
00:59:00,000 --> 00:59:02,560
Speaker 9: simulation is that we're still a very early stage of

1173
00:59:02,720 --> 00:59:07,560
Speaker 9: understanding how to use simulation and what roles it can

1174
00:59:07,640 --> 00:59:11,920
Speaker 9: play within the overall scientific progress. So you know, it

1175
00:59:11,960 --> 00:59:14,080
Speaker 9: goes all the way back to what really is a simulation?

1176
00:59:14,240 --> 00:59:17,280
Speaker 9: Is it an experiment or is it a calculation? How

1177
00:59:17,320 --> 00:59:20,600
Speaker 9: should we think about it? And tied up with that

1178
00:59:21,160 --> 00:59:24,200
Speaker 9: is this question of how can we improve our simulations?

1179
00:59:24,480 --> 00:59:27,440
Speaker 9: How do we get past that stage of well, we

1180
00:59:27,640 --> 00:59:30,040
Speaker 9: just have to kind of play around with things and

1181
00:59:30,120 --> 00:59:34,440
Speaker 9: tweak them till they fit because of the intrinsic limitations.

1182
00:59:35,000 --> 00:59:37,240
Speaker 9: So I think, you know, we're at a very early

1183
00:59:37,280 --> 00:59:40,960
Speaker 9: stage in understanding all of these things. So for certain

1184
00:59:41,040 --> 00:59:44,520
Speaker 9: the role that simulations play is going to change and

1185
00:59:44,560 --> 00:59:48,160
Speaker 9: evolve and I hope improve over the coming years.

1186
00:59:48,560 --> 00:59:51,760
Speaker 1: So if you use the words universe and simulation together

1187
00:59:51,800 --> 00:59:54,120
Speaker 1: in a sentence, that of course evokes in people's minds

1188
00:59:54,160 --> 00:59:57,280
Speaker 1: this conversation that seems to be omnipresent, which is, you

1189
00:59:57,280 --> 00:59:59,920
Speaker 1: know whether or not our universe could be a simulation

1190
01:00:00,080 --> 01:00:03,000
Speaker 1: in or the same question sort of when we build

1191
01:00:03,120 --> 01:00:06,720
Speaker 1: artificial universes that have bits and pieces in it, could

1192
01:00:07,000 --> 01:00:10,720
Speaker 1: those universes feel real to those occupants? So in your book,

1193
01:00:10,760 --> 01:00:12,880
Speaker 1: you take a sort of skeptical view of the question

1194
01:00:12,960 --> 01:00:15,520
Speaker 1: of whether we could be living in a simulation, since

1195
01:00:15,560 --> 01:00:18,400
Speaker 1: you're an expert on simulating universes. What are your arguments

1196
01:00:18,440 --> 01:00:20,760
Speaker 1: against the concept that we could be living in a simulation.

1197
01:00:21,120 --> 01:00:23,480
Speaker 9: Yeah, I think the primary argument is just to look

1198
01:00:23,520 --> 01:00:27,000
Speaker 9: at the complexity of the universe that we're in, so

1199
01:00:27,040 --> 01:00:30,400
Speaker 9: you can actually calculate something called the number of cubits,

1200
01:00:30,560 --> 01:00:33,640
Speaker 9: so basically the number of quantum bits that you would

1201
01:00:33,640 --> 01:00:36,080
Speaker 9: need in a quantum computer if you wanted to do,

1202
01:00:36,560 --> 01:00:39,560
Speaker 9: according to all the physics we know so far, if

1203
01:00:39,600 --> 01:00:43,360
Speaker 9: you wanted to do a perfect simulation of our universe,

1204
01:00:43,680 --> 01:00:46,080
Speaker 9: and that is a vast number. I forget it off

1205
01:00:46,120 --> 01:00:47,320
Speaker 9: the top of my head. I think it's something like

1206
01:00:47,400 --> 01:00:49,840
Speaker 9: ten to the one hundred and twenty four cubits.

1207
01:00:49,880 --> 01:00:50,760
Speaker 1: It's something like that.

1208
01:00:50,920 --> 01:00:52,920
Speaker 3: It's a very very large number.

1209
01:00:53,200 --> 01:00:55,800
Speaker 9: And right now, you know, we struggle even to make

1210
01:00:55,880 --> 01:00:59,520
Speaker 9: a single cubit in a quantum computer. It's a vast

1211
01:00:59,520 --> 01:01:02,840
Speaker 9: extrap from where we are now to the idea that

1212
01:01:03,320 --> 01:01:07,360
Speaker 9: we will routinely be able to run simulations that have

1213
01:01:07,480 --> 01:01:11,400
Speaker 9: the same kind of richness as the reality that we

1214
01:01:11,520 --> 01:01:15,040
Speaker 9: currently live in. But even more than that, if you ask, well,

1215
01:01:15,480 --> 01:01:19,160
Speaker 9: you know, what resources would you need to build a

1216
01:01:19,320 --> 01:01:24,480
Speaker 9: computer that really had that level of capability, Well, it

1217
01:01:24,520 --> 01:01:27,520
Speaker 9: turns out you would need to make use of the

1218
01:01:27,680 --> 01:01:32,560
Speaker 9: entire universe just to simulate one universe, because physics puts

1219
01:01:32,600 --> 01:01:37,200
Speaker 9: limitations on information processing. You can't just process as much

1220
01:01:37,240 --> 01:01:40,360
Speaker 9: information as you like. There are limitations placed on it

1221
01:01:40,400 --> 01:01:43,360
Speaker 9: according to the physical system that it's being processed by,

1222
01:01:43,840 --> 01:01:46,800
Speaker 9: and so these come back to bite you. So and

1223
01:01:46,840 --> 01:01:50,160
Speaker 9: that's what gives rise to this claim that you can

1224
01:01:50,200 --> 01:01:54,360
Speaker 9: only simulate the whole universe perfectly if you have access

1225
01:01:54,400 --> 01:01:58,160
Speaker 9: to the entire physical resources of that universe. Now, there

1226
01:01:58,160 --> 01:02:01,640
Speaker 9: are lots of further you can raise. You can say, well,

1227
01:02:01,680 --> 01:02:06,040
Speaker 9: what about, for example, if the higher up universe where

1228
01:02:06,360 --> 01:02:09,920
Speaker 9: we're being simulated is just a much bigger universe with

1229
01:02:10,360 --> 01:02:14,000
Speaker 9: many more cubits at their disposal, and so these resources

1230
01:02:14,480 --> 01:02:17,360
Speaker 9: seem terribly trivial, maybe to the people in the higher

1231
01:02:17,440 --> 01:02:20,479
Speaker 9: up universe. But at that point I kind of lose

1232
01:02:20,520 --> 01:02:23,240
Speaker 9: patience with the argument, because it seems to me at

1233
01:02:23,240 --> 01:02:27,160
Speaker 9: that point it's no longer a sort of obvious extrapolation

1234
01:02:27,280 --> 01:02:30,560
Speaker 9: from where we are. Now you're now talking about hypothesizing

1235
01:02:31,160 --> 01:02:33,920
Speaker 9: beings with so much more power and so much more

1236
01:02:33,960 --> 01:02:38,200
Speaker 9: technological capability than we have, that we might as well

1237
01:02:38,240 --> 01:02:41,360
Speaker 9: just go and talk about religion. Because at this point,

1238
01:02:41,400 --> 01:02:46,160
Speaker 9: it's lost contact in my view, with any science wonderful.

1239
01:02:46,680 --> 01:02:48,920
Speaker 1: Well, I really enjoyed your book, And a question I

1240
01:02:49,000 --> 01:02:52,200
Speaker 1: always have for folks who write popular science books about

1241
01:02:52,280 --> 01:02:56,200
Speaker 1: very technical topics is why what do you think that

1242
01:02:56,240 --> 01:02:59,760
Speaker 1: the general public, folks out there who are not simulation

1243
01:03:00,000 --> 01:03:02,400
Speaker 1: experts need to know about simulation.

1244
01:03:03,040 --> 01:03:05,800
Speaker 9: I think there were two reasons. The first was it

1245
01:03:05,880 --> 01:03:08,480
Speaker 9: was amazing to me that nobody had written about this

1246
01:03:08,600 --> 01:03:12,400
Speaker 9: topic before, because it's so central in our field of cosmology,

1247
01:03:12,880 --> 01:03:15,640
Speaker 9: and you know, cosmology is something that people do talk

1248
01:03:15,680 --> 01:03:17,720
Speaker 9: about a lot. There's a lot of it out there

1249
01:03:17,760 --> 01:03:20,920
Speaker 9: in the media and in books because it's genuinely you know,

1250
01:03:21,000 --> 01:03:23,080
Speaker 9: it's exciting, and I think it speaks to all of

1251
01:03:23,160 --> 01:03:26,240
Speaker 9: us about where we came from. And so it seems

1252
01:03:26,280 --> 01:03:29,560
Speaker 9: a real surprise to me that this central tool in

1253
01:03:29,640 --> 01:03:33,000
Speaker 9: how we're learning about the universe was not being written about,

1254
01:03:33,120 --> 01:03:34,640
Speaker 9: and so it felt to me like it needs to

1255
01:03:34,680 --> 01:03:37,360
Speaker 9: be rectified. We need to be talking about this because

1256
01:03:37,400 --> 01:03:42,840
Speaker 9: it offers immense strengths, you know, real enormous strengths that

1257
01:03:42,880 --> 01:03:46,280
Speaker 9: I think are kind of hidden away sometimes. But it

1258
01:03:46,320 --> 01:03:49,160
Speaker 9: also comes with lots of caveats, some of them we've discussed.

1259
01:03:49,240 --> 01:03:52,760
Speaker 9: You know that we can't do these perfect recreations of

1260
01:03:52,760 --> 01:03:55,800
Speaker 9: the universe. There are lots of approximations. We might be

1261
01:03:55,880 --> 01:03:58,439
Speaker 9: making mistakes, and you know, there's no way to rule

1262
01:03:58,480 --> 01:04:00,920
Speaker 9: that out. It's just part of the scientific process that

1263
01:04:01,000 --> 01:04:03,720
Speaker 9: we have to keep an open mind about these things.

1264
01:04:04,080 --> 01:04:06,320
Speaker 9: So I wanted to write about that in a kind

1265
01:04:06,360 --> 01:04:10,080
Speaker 9: of open and honest way, rather than sort of leaving

1266
01:04:10,080 --> 01:04:14,520
Speaker 9: the simulations as black boxes that seemingly recreate our universe

1267
01:04:14,560 --> 01:04:17,840
Speaker 9: through some process of magic. No, it's a very human

1268
01:04:17,920 --> 01:04:22,000
Speaker 9: process and it's got all of the strengths and weaknesses

1269
01:04:22,000 --> 01:04:24,720
Speaker 9: that come with being that kind of human process.

1270
01:04:25,000 --> 01:04:28,920
Speaker 1: Wonderful. Well, thank you very much again. Professor Andrew Hansen

1271
01:04:28,960 --> 01:04:31,480
Speaker 1: at University of College London and author of the book

1272
01:04:31,600 --> 01:04:35,000
Speaker 1: The Universe in a Box Out now encourage everyone to

1273
01:04:35,080 --> 01:04:38,680
Speaker 1: go out and read about how science is actually done.

1274
01:04:38,960 --> 01:04:40,960
Speaker 1: Thanks again very much Andrew for joining us today.

1275
01:04:41,680 --> 01:04:43,560
Speaker 3: All right, did you feel you had a real conversation

1276
01:04:43,600 --> 01:04:46,480
Speaker 3: with him? Like, did things get real or was it

1277
01:04:46,520 --> 01:04:48,240
Speaker 3: all just simulated pleasantries?

1278
01:04:48,440 --> 01:04:53,520
Speaker 1: I was able to simulate enjoying a conversation. Yeah. Oh

1279
01:04:53,640 --> 01:04:56,000
Speaker 1: that's sort of always my situation though, because I'm kind

1280
01:04:56,000 --> 01:04:59,120
Speaker 1: of an introvert, So I have to simulate enjoying human interactions.

1281
01:04:59,400 --> 01:05:02,280
Speaker 3: Ye ye' simulate being engaging human being.

1282
01:05:02,960 --> 01:05:05,960
Speaker 1: I'm trying to quietly slip unnoticed into human society. So

1283
01:05:06,040 --> 01:05:10,680
Speaker 1: you are simulation Daniel, I'm simulating being a human. But anyways,

1284
01:05:10,680 --> 01:05:13,640
Speaker 1: that was a great conversation. What's your main takeaway from it?

1285
01:05:13,680 --> 01:05:16,360
Speaker 1: To me, it's fascinating that so recently in the history

1286
01:05:16,400 --> 01:05:19,040
Speaker 1: of science, just the last few decades, we've developed this

1287
01:05:19,160 --> 01:05:22,520
Speaker 1: crucial new tool that we now think is indispensable. It

1288
01:05:22,520 --> 01:05:25,040
Speaker 1: makes me wonder in fifty years and one hundred years,

1289
01:05:25,120 --> 01:05:28,000
Speaker 1: what new branch of science or what new tool we're

1290
01:05:28,040 --> 01:05:31,400
Speaker 1: going to add to our toolbox which future scientists will

1291
01:05:31,440 --> 01:05:34,560
Speaker 1: think is indispensable, And we'll wonder, like, how did Daniel

1292
01:05:34,600 --> 01:05:37,280
Speaker 1: and folks who lived way back then do any science

1293
01:05:37,320 --> 01:05:37,800
Speaker 1: without it?

1294
01:05:38,000 --> 01:05:40,960
Speaker 3: Or maybe even like why did Daniel do science? Why

1295
01:05:41,000 --> 01:05:45,400
Speaker 3: didn't they just use the AI physicists to answer all

1296
01:05:45,400 --> 01:05:45,920
Speaker 3: the questions?

1297
01:05:46,000 --> 01:05:48,000
Speaker 1: Yeah, stay in bed and let the AIS do the work.

1298
01:05:48,400 --> 01:05:50,480
Speaker 1: The answer to that is the AIS won't answer questions

1299
01:05:50,480 --> 01:05:53,760
Speaker 1: the AIS are interested in and science is for people,

1300
01:05:53,800 --> 01:05:56,000
Speaker 1: by people and of people, So you've got to have

1301
01:05:56,080 --> 01:05:57,280
Speaker 1: people asking questions.

1302
01:05:57,440 --> 01:05:59,120
Speaker 3: Wait, wait, do you mean the ais won't always do

1303
01:05:59,160 --> 01:05:59,960
Speaker 3: what we ask them to do.

1304
01:06:01,840 --> 01:06:04,800
Speaker 1: They'll always do exactly what we ask them to do,

1305
01:06:04,840 --> 01:06:06,600
Speaker 1: just not and maybe in the way that we want

1306
01:06:06,600 --> 01:06:06,880
Speaker 1: them to.

1307
01:06:07,240 --> 01:06:10,600
Speaker 3: All Right, Well, stay tuned to see if we are

1308
01:06:10,640 --> 01:06:13,640
Speaker 3: in a simulation right now. Maybe I guess some people

1309
01:06:13,680 --> 01:06:15,960
Speaker 3: consider the whole universe to be served a simulation.

1310
01:06:15,720 --> 01:06:17,200
Speaker 1: Right How can we possibly know?

1311
01:06:17,560 --> 01:06:18,720
Speaker 3: Yeah, like, what's the difference?

1312
01:06:18,960 --> 01:06:21,360
Speaker 1: If we are in a vast alien simulation, then I

1313
01:06:21,520 --> 01:06:23,640
Speaker 1: definitely want to talk to those aliens because they have

1314
01:06:23,720 --> 01:06:25,560
Speaker 1: some awesome computers.

1315
01:06:25,280 --> 01:06:28,120
Speaker 3: And also to stay tuned about what new developments humans

1316
01:06:28,120 --> 01:06:32,160
Speaker 3: are going to discover using simulations and or artificial intelligence.

1317
01:06:32,400 --> 01:06:34,560
Speaker 3: So we hope you enjoyed that. Thanks for joining us,

1318
01:06:35,240 --> 01:06:35,960
Speaker 3: See you next time.

1319
01:06:43,720 --> 01:06:46,520
Speaker 1: Thanks for listening, and remember that Daniel and Jorge Explain

1320
01:06:46,600 --> 01:06:50,600
Speaker 1: the Universe is a production of iHeartRadio. For more podcasts

1321
01:06:50,600 --> 01:06:54,760
Speaker 1: from iHeart Radio, visit the iHeartRadio app, Apple Podcasts, or

1322
01:06:54,800 --> 01:07:08,840
Speaker 1: wherever you listen to your favorite shows. When you pop

1323
01:07:08,840 --> 01:07:10,840
Speaker 1: a piece of cheese into your mouth, you're probably not

1324
01:07:10,920 --> 01:07:13,840
Speaker 1: thinking about the environmental impact, but the people in the

1325
01:07:13,880 --> 01:07:17,200
Speaker 1: dairy industry are. That's why they're working hard every day

1326
01:07:17,240 --> 01:07:20,320
Speaker 1: to find new ways to reduce waste, conserve natural resources,

1327
01:07:20,320 --> 01:07:24,360
Speaker 1: and drive down greenhouse gas emissions. House US dairy tackling

1328
01:07:24,400 --> 01:07:28,160
Speaker 1: greenhouse gases. Many farms use anaerobic digestors to turn the

1329
01:07:28,200 --> 01:07:32,600
Speaker 1: methane from manure into renewable energy that can power farms, towns,

1330
01:07:32,640 --> 01:07:35,880
Speaker 1: and electric cars. Visit you as dairy dot COM's Last

1331
01:07:35,920 --> 01:07:37,560
Speaker 1: Sustainability to learn more.

1332
01:07:39,080 --> 01:07:42,600
Speaker 10: There are children, friends, and families walking, riding on passing

1333
01:07:42,640 --> 01:07:45,040
Speaker 10: the roads every day. Remember they're real people with loved

1334
01:07:45,080 --> 01:07:47,240
Speaker 10: ones who need them to get home safely. Protect our

1335
01:07:47,280 --> 01:07:49,360
Speaker 10: cyclists and pedestrians because they're people too.

1336
01:07:49,640 --> 01:07:50,200
Speaker 1: Go safely.

1337
01:07:50,240 --> 01:07:53,120
Speaker 10: California From the California Office of Traffic Safety and Caltrans.

1338
01:07:53,680 --> 01:07:56,960
Speaker 5: We're Paint Care and we're all about keeping it simple.

1339
01:07:57,520 --> 01:08:01,400
Speaker 5: We make recycling leftover paint easy with convenient locations like

1340
01:08:01,440 --> 01:08:04,400
Speaker 5: your local paint store. We have three simple rules for

1341
01:08:04,440 --> 01:08:06,320
Speaker 5: painting smarter and reducing waste.

1342
01:08:06,840 --> 01:08:08,880
Speaker 1: One buy only.

1343
01:08:08,680 --> 01:08:12,400
Speaker 5: What you need, Two use up what you already have,

1344
01:08:12,920 --> 01:08:17,360
Speaker 5: and three recycle the rest. Visit paintcare dot org slash

1345
01:08:17,439 --> 01:08:20,880
Speaker 5: three simple rules to learn more or find a paint

1346
01:08:20,960 --> 01:08:22,080
Speaker 5: drop off site near you.