1
00:00:15,316 --> 00:00:22,916
Speaker 1: Pushkin. I'm looking right now at the Google Maps satellite

2
00:00:22,916 --> 00:00:25,556
Speaker 1: image of my block, of my house and the houses

3
00:00:25,796 --> 00:00:28,116
Speaker 1: next to my house, and I can tell from looking

4
00:00:28,116 --> 00:00:29,996
Speaker 1: at the picture that it is at least a year

5
00:00:30,036 --> 00:00:33,076
Speaker 1: or so old. I know this because last summer, my

6
00:00:33,156 --> 00:00:35,636
Speaker 1: kids and I watched as our next door neighbors had

7
00:00:35,676 --> 00:00:38,396
Speaker 1: the time of their lives in their brand new above

8
00:00:38,476 --> 00:00:42,036
Speaker 1: ground pool, and in this satellite picture on Google Maps,

9
00:00:42,556 --> 00:00:45,636
Speaker 1: there is no pool. This is fine. I don't need

10
00:00:45,676 --> 00:00:48,156
Speaker 1: to look at Google Maps to know that my neighbors

11
00:00:48,156 --> 00:00:50,676
Speaker 1: are living their best lives now, and I perhaps am not.

12
00:00:51,436 --> 00:00:55,956
Speaker 1: But there are other instances, other use cases, if you will,

13
00:00:56,236 --> 00:00:59,396
Speaker 1: when you might want satellite images that are updated more

14
00:00:59,436 --> 00:01:03,396
Speaker 1: frequently than every year or so. For example, if you

15
00:01:03,476 --> 00:01:05,916
Speaker 1: were a hedge fund who wanted to know how many

16
00:01:05,996 --> 00:01:08,916
Speaker 1: cars were in the parking lot of every home depot

17
00:01:08,956 --> 00:01:13,356
Speaker 1: in America last weekend, or an NGO monitoring deforestation in

18
00:01:13,396 --> 00:01:16,396
Speaker 1: the Amazon, or a government that wanted to see if

19
00:01:16,436 --> 00:01:19,476
Speaker 1: foreign troops were massing near your border. If you were

20
00:01:19,516 --> 00:01:22,596
Speaker 1: any of those things, you might want satellite images that

21
00:01:22,636 --> 00:01:25,996
Speaker 1: are updated much more frequently than once a year. You

22
00:01:26,076 --> 00:01:31,916
Speaker 1: might even want a new picture from space every single day.

23
00:01:33,996 --> 00:01:36,916
Speaker 1: I'm Jacob Goldstein and this is What's Your Problem, the

24
00:01:36,956 --> 00:01:39,876
Speaker 1: show where I talked to entrepreneurs and engineers about how

25
00:01:39,876 --> 00:01:42,196
Speaker 1: they're going to change the world once they solve a

26
00:01:42,236 --> 00:01:47,596
Speaker 1: few problems. My guest today is Will Marshall, the co

27
00:01:47,716 --> 00:01:51,676
Speaker 1: founder and CEO of Planet. Planet is a private company

28
00:01:51,916 --> 00:01:54,716
Speaker 1: and they own a fleet of over two hundred satellites

29
00:01:54,876 --> 00:01:58,436
Speaker 1: that takes a picture of the entire Earth every day.

30
00:01:58,916 --> 00:02:01,996
Speaker 1: Will's current problem how to turn a daily picture of

31
00:02:02,036 --> 00:02:06,316
Speaker 1: the Earth into useful information. Will started his career as

32
00:02:06,356 --> 00:02:09,196
Speaker 1: an engineer at NASA, and while he was there in

33
00:02:09,236 --> 00:02:11,716
Speaker 1: a colleague came up with the idea for Planet. But

34
00:02:11,756 --> 00:02:16,476
Speaker 1: to get started, they faced a simple, massive problem. Satellites

35
00:02:16,556 --> 00:02:21,356
Speaker 1: were way, way, way too expensive, about a billion dollars

36
00:02:21,436 --> 00:02:24,316
Speaker 1: per satellite, and they knew they'd never be able to

37
00:02:24,316 --> 00:02:27,116
Speaker 1: get a whole fleet of satellites into space at that price.

38
00:02:27,436 --> 00:02:30,276
Speaker 1: So they asked themselves a question, what's something that is

39
00:02:30,316 --> 00:02:34,236
Speaker 1: a lot like a satellite, but much cheaper. We had

40
00:02:34,556 --> 00:02:38,196
Speaker 1: looked at a phone and said, well, that's got most

41
00:02:38,476 --> 00:02:41,996
Speaker 1: of the ingredients of a satellite. If you look at

42
00:02:41,996 --> 00:02:45,196
Speaker 1: a smartphone, it has GPS telling you where you are,

43
00:02:45,276 --> 00:02:49,396
Speaker 1: It has cameras, It has hard drives, it has accelerometers,

44
00:02:49,556 --> 00:02:52,356
Speaker 1: It has rate gyros that used for gaming, so people

45
00:02:52,396 --> 00:02:54,636
Speaker 1: know when they're tipping the thing. Actually, that's kind of

46
00:02:54,676 --> 00:02:56,996
Speaker 1: handy for spacecraft. The only thing it doesn't have that

47
00:02:57,076 --> 00:03:00,556
Speaker 1: spacecraft has is continuous power supply. It has a battery, right,

48
00:03:00,596 --> 00:03:03,516
Speaker 1: but it doesn't have the ability to power itself. So

49
00:03:04,396 --> 00:03:07,196
Speaker 1: and it doesn't have a bright enough radio if you like.

50
00:03:07,316 --> 00:03:09,796
Speaker 1: It's it's radio is the most short range, whereas we

51
00:03:09,836 --> 00:03:13,676
Speaker 1: need a more substantive radio. But it has a lot

52
00:03:13,716 --> 00:03:17,276
Speaker 1: of ingredients. But also it doesn't cost a billion dollars

53
00:03:17,356 --> 00:03:19,476
Speaker 1: on the plus, well that's the other thing. Yeah, so

54
00:03:19,516 --> 00:03:22,316
Speaker 1: because maybe because five hundred dollars not five hundred million dollars.

55
00:03:22,516 --> 00:03:25,716
Speaker 1: And so we thought it would be cunning to throw

56
00:03:25,756 --> 00:03:28,156
Speaker 1: some of these phones in space and see what happened

57
00:03:28,196 --> 00:03:30,156
Speaker 1: to them. And we did. This is when we're as

58
00:03:30,636 --> 00:03:32,676
Speaker 1: you like, put it in a box with the solar

59
00:03:32,716 --> 00:03:36,716
Speaker 1: panel and a stronger radio, yeah exactly, binger, and then

60
00:03:36,756 --> 00:03:39,716
Speaker 1: put it into space. They were tumbling around and they

61
00:03:39,716 --> 00:03:42,596
Speaker 1: took a few pictures of the Earth with the cameras

62
00:03:42,596 --> 00:03:45,236
Speaker 1: on the phone, just with their regular smartphone here. Yeah,

63
00:03:45,436 --> 00:03:48,956
Speaker 1: and then we had amateur radio observers pick up the

64
00:03:49,036 --> 00:03:53,476
Speaker 1: signals from these, they're like texting the selfie of the

65
00:03:53,516 --> 00:03:56,836
Speaker 1: Earth batter Yes, And then we recompiled all the bits

66
00:03:56,836 --> 00:03:59,116
Speaker 1: of the pictures into a picture of the Earth. And

67
00:03:59,116 --> 00:04:01,196
Speaker 1: then then we were like, oh, okay and dipso did

68
00:04:01,196 --> 00:04:03,636
Speaker 1: it work? Yes, it worked. The phones worked just fine

69
00:04:03,676 --> 00:04:07,196
Speaker 1: in space. We flew three of them, three Google Nexus ones,

70
00:04:07,556 --> 00:04:09,716
Speaker 1: because they were easier to hack at than the iPhone.

71
00:04:09,916 --> 00:04:11,996
Speaker 1: And that was when we decided, okay, we could do

72
00:04:12,036 --> 00:04:14,516
Speaker 1: this very differently, but it's probably makes more sense as

73
00:04:14,556 --> 00:04:18,436
Speaker 1: a private company because NASA set about doing exploration and science,

74
00:04:18,556 --> 00:04:21,516
Speaker 1: but this was more applied. This is more about commercial

75
00:04:21,956 --> 00:04:26,596
Speaker 1: benefits or humanitarian benefits or it's very applied in nature.

76
00:04:26,636 --> 00:04:28,596
Speaker 1: And so we thought we could do it a little

77
00:04:28,596 --> 00:04:31,196
Speaker 1: bit faster outside of NASA. So we left NASA to

78
00:04:31,236 --> 00:04:34,036
Speaker 1: do this project of setting out this goal of launching

79
00:04:34,036 --> 00:04:36,276
Speaker 1: about one hundred satellites to image the whole Earth every

80
00:04:36,276 --> 00:04:40,396
Speaker 1: single day. So you leave NASA at this moment when

81
00:04:40,716 --> 00:04:43,796
Speaker 1: a satellite costs a billion dollars, so you need something

82
00:04:43,836 --> 00:04:47,876
Speaker 1: between a five hundred dollars smartphone and a billion dollars satellite, right,

83
00:04:47,916 --> 00:04:49,636
Speaker 1: So what is the thing you've got to build, like

84
00:04:50,196 --> 00:04:52,116
Speaker 1: more or less, how much does it have to cost?

85
00:04:52,156 --> 00:04:53,716
Speaker 1: How cheap do you have to be able to make

86
00:04:53,716 --> 00:04:55,356
Speaker 1: a satellite when you set out for it to work,

87
00:04:55,436 --> 00:04:58,676
Speaker 1: we thought that we could do it of order a

88
00:04:58,756 --> 00:05:01,676
Speaker 1: couple of hundred thousand dollars per satellite, and so to

89
00:05:01,836 --> 00:05:05,596
Speaker 1: get from a billion to three hundred thousand all the

90
00:05:05,596 --> 00:05:09,476
Speaker 1: way down, Like, can you just tell me maybe one

91
00:05:10,876 --> 00:05:14,276
Speaker 1: problem you solved, like one thing you had to figure out? Obviously,

92
00:05:14,556 --> 00:05:16,996
Speaker 1: I'm sure you had to figure out a thousand things,

93
00:05:17,036 --> 00:05:19,236
Speaker 1: a billion things to save a billion dollars, but like,

94
00:05:19,316 --> 00:05:22,196
Speaker 1: what's one little problem you solved on their way? One

95
00:05:22,356 --> 00:05:26,556
Speaker 1: is radios. So the smallest radio that would work in

96
00:05:26,556 --> 00:05:28,956
Speaker 1: the frequency band that is designated for satellites to do

97
00:05:28,996 --> 00:05:32,316
Speaker 1: Earth imaging costs more than a million dollars I think

98
00:05:32,356 --> 00:05:35,676
Speaker 1: a few million dollars and was bigger than our entire

99
00:05:35,756 --> 00:05:37,996
Speaker 1: satellite is meant to be. And we were like, well,

100
00:05:38,116 --> 00:05:39,716
Speaker 1: damn it, that's not going to work because most of

101
00:05:39,716 --> 00:05:41,956
Speaker 1: the satellite needs to be our telescope. We need to

102
00:05:41,996 --> 00:05:44,876
Speaker 1: scram it into a little motherboard and have a little

103
00:05:44,876 --> 00:05:48,276
Speaker 1: antenna on the side. And by the way, that that

104
00:05:48,316 --> 00:05:51,596
Speaker 1: system would only produce about a few megabits a second,

105
00:05:51,836 --> 00:05:56,036
Speaker 1: and we needed gigabits a second, so it was not

106
00:05:56,156 --> 00:05:58,596
Speaker 1: only too expensive and too massive, it didn't even do

107
00:05:58,636 --> 00:06:01,636
Speaker 1: the job. And so we built our own radios from scratch,

108
00:06:01,836 --> 00:06:04,836
Speaker 1: designed these antenna that was based on pieces of wire,

109
00:06:04,876 --> 00:06:07,636
Speaker 1: but their machine learning designs so that they even though

110
00:06:07,636 --> 00:06:10,836
Speaker 1: it's just like a paper clip, almost we now get

111
00:06:11,196 --> 00:06:13,876
Speaker 1: it's the perfect paper clip. Yes, it's a perfect dedesigned

112
00:06:13,876 --> 00:06:17,596
Speaker 1: paper clip. And we can now get one point eight

113
00:06:17,676 --> 00:06:20,996
Speaker 1: gigabits a second over a thousand kilometer arrange, which is

114
00:06:21,076 --> 00:06:23,916
Speaker 1: unheard of. And how much does it cost? How much

115
00:06:23,956 --> 00:06:26,836
Speaker 1: does it cost of order a thousand dollars thousand dollars

116
00:06:26,876 --> 00:06:30,756
Speaker 1: instead of a million, Yeah, exactly. You know, we didn't

117
00:06:30,796 --> 00:06:33,916
Speaker 1: take space grade components. We took components that we knew

118
00:06:33,956 --> 00:06:36,876
Speaker 1: of work in a vacuum and tried them and had

119
00:06:37,036 --> 00:06:40,476
Speaker 1: instead of trying to put all our energies into one spacecraft,

120
00:06:40,556 --> 00:06:42,356
Speaker 1: we would build lots of them and if one or

121
00:06:42,396 --> 00:06:45,276
Speaker 1: two spacecraft die, it's not the end of the world either.

122
00:06:45,436 --> 00:06:49,396
Speaker 1: So by doing that it enabled us to leverage the

123
00:06:49,516 --> 00:06:52,116
Speaker 1: latest technologies. Because if you don't care so much, if

124
00:06:52,156 --> 00:06:54,916
Speaker 1: it's for sure going to work. You know, you just

125
00:06:54,956 --> 00:06:57,596
Speaker 1: want it to be like eighty percent probable of working

126
00:06:57,676 --> 00:07:00,716
Speaker 1: or ninety percent probable of working. You can take bigger risk,

127
00:07:00,756 --> 00:07:03,836
Speaker 1: which actually means you can more use more latest gadgets tree. Yeah,

128
00:07:03,876 --> 00:07:06,476
Speaker 1: so that's a bigger idea. So by using cheaper components,

129
00:07:07,116 --> 00:07:10,436
Speaker 1: it's okay if they fail, because you're launching one hundred

130
00:07:10,476 --> 00:07:15,076
Speaker 1: satellites and if ninety five of them naked, that's just fun.

131
00:07:15,436 --> 00:07:17,116
Speaker 1: That's just fine. And let me give you a concrete

132
00:07:17,116 --> 00:07:19,156
Speaker 1: example of the last launch we did just a few

133
00:07:19,156 --> 00:07:23,356
Speaker 1: weeks ago on a SpaceX rocket. We launched forty four

134
00:07:23,476 --> 00:07:27,716
Speaker 1: satellites and one of them hasn't worked, and we barely

135
00:07:28,596 --> 00:07:31,796
Speaker 1: you know, blinked as an organization, you know it was.

136
00:07:32,236 --> 00:07:35,076
Speaker 1: In fact, that's part of the place they're they're built

137
00:07:35,076 --> 00:07:38,476
Speaker 1: to fail. What how is there some amount of time

138
00:07:38,596 --> 00:07:41,116
Speaker 1: like you expect ten percent failure rate or five percent

139
00:07:41,196 --> 00:07:42,916
Speaker 1: or do you have that sort of penciled down? You know,

140
00:07:42,996 --> 00:07:45,716
Speaker 1: I've I've asked the team to shoot for more like

141
00:07:45,796 --> 00:07:48,916
Speaker 1: ten percent or twenty percent failure rate, but we've never

142
00:07:48,956 --> 00:07:52,676
Speaker 1: achieved that. We've always been much lower. You need to

143
00:07:52,716 --> 00:07:56,356
Speaker 1: take more risks, you need to make the right or

144
00:07:56,396 --> 00:07:59,476
Speaker 1: whatever right, right, right right. That's really just that we're

145
00:07:59,516 --> 00:08:02,796
Speaker 1: being too conservative. So you build the satellites. It works.

146
00:08:02,956 --> 00:08:05,356
Speaker 1: You got your cheap satellites. Now you got to get

147
00:08:05,396 --> 00:08:10,156
Speaker 1: them to space. Yes, correct. We call it piggybacking, piggyback payloads.

148
00:08:10,796 --> 00:08:13,156
Speaker 1: So obviously there's a big satellite paying for most of

149
00:08:13,156 --> 00:08:15,236
Speaker 1: the cust of the bucket, and we say, well, there's

150
00:08:15,236 --> 00:08:16,756
Speaker 1: a little bit of space there, can we shove a

151
00:08:16,756 --> 00:08:19,556
Speaker 1: few satellites in. It's like we're hitchhikers. So it's like

152
00:08:19,556 --> 00:08:21,516
Speaker 1: we're going to the Florida and we're sticking up our

153
00:08:21,516 --> 00:08:23,636
Speaker 1: thumb and saying, hey, can we can we come up

154
00:08:23,636 --> 00:08:26,996
Speaker 1: with you guys? Well, and your satellites are in fact small, right,

155
00:08:27,076 --> 00:08:29,596
Speaker 1: just like yeah, very small. Tell me what one looks like.

156
00:08:29,716 --> 00:08:31,236
Speaker 1: It is about the size of a loafer bread is

157
00:08:31,276 --> 00:08:35,396
Speaker 1: ten by ten by thirty centimeters. It weighs about five kilograms,

158
00:08:35,676 --> 00:08:38,796
Speaker 1: So it's so tiny for a satellite, tiny, smaller than

159
00:08:39,196 --> 00:08:42,476
Speaker 1: anybody would think of a satellite being absolutely, you can

160
00:08:42,516 --> 00:08:44,396
Speaker 1: take them on our hand luggage. In fact, that's how

161
00:08:44,436 --> 00:08:46,236
Speaker 1: we did get them to the launch site. Early on.

162
00:08:46,276 --> 00:08:48,596
Speaker 1: You checked them on the you carried out, just carried

163
00:08:48,636 --> 00:08:50,596
Speaker 1: them on a carry on luggage. Yeah, you put them

164
00:08:50,596 --> 00:08:52,916
Speaker 1: in the overhead or just put them in the overhead.

165
00:08:53,076 --> 00:08:54,916
Speaker 1: I mean we wrap them up a bit and stuff,

166
00:08:54,916 --> 00:08:57,436
Speaker 1: and we have to tell you really have carried them

167
00:08:57,436 --> 00:09:00,516
Speaker 1: on the plane. Oh yeah, yeah to the launch site. Yeah.

168
00:09:00,556 --> 00:09:05,276
Speaker 1: So I mean in a sense like the satellites are cool,

169
00:09:06,556 --> 00:09:10,076
Speaker 1: but you're a data business, right, Your business is not

170
00:09:10,276 --> 00:09:13,476
Speaker 1: really launching satellites. It's selling data and specific So you

171
00:09:13,516 --> 00:09:17,636
Speaker 1: take a picture of the whole earth every day. Who

172
00:09:17,676 --> 00:09:19,036
Speaker 1: do you sell it to? What do you do with it?

173
00:09:19,356 --> 00:09:21,636
Speaker 1: So the biggest one is agriculture, the one I just mentioned,

174
00:09:21,636 --> 00:09:24,756
Speaker 1: and we help them improve efficiency by twenty or forty percent,

175
00:09:25,076 --> 00:09:28,396
Speaker 1: decrease use of fertilizer, similar sort of amounts. So that's

176
00:09:28,396 --> 00:09:32,796
Speaker 1: a big deal for big industry like that forestry. We

177
00:09:32,836 --> 00:09:37,876
Speaker 1: work with mapping like Google, help improve the maps, civil

178
00:09:37,916 --> 00:09:41,836
Speaker 1: government with disaster response, so floods, fires, earthquakes. We help

179
00:09:42,236 --> 00:09:46,276
Speaker 1: countries monitor marine protected areas and coral reef zones and

180
00:09:46,316 --> 00:09:51,596
Speaker 1: make sure that there's not illegal fishing. Insurance, hedge funds.

181
00:09:52,156 --> 00:09:56,596
Speaker 1: Spying comes to mind, like our intelligence agencies clients of yours,

182
00:09:56,676 --> 00:10:00,516
Speaker 1: Yes they are, but we take a super important principle

183
00:10:00,556 --> 00:10:03,796
Speaker 1: which is why I think it's different than the classical

184
00:10:03,876 --> 00:10:07,396
Speaker 1: way the earth imaging can be used in this arena,

185
00:10:07,476 --> 00:10:11,556
Speaker 1: which is non exclusivity. We can provide that data because

186
00:10:11,556 --> 00:10:14,356
Speaker 1: we scan automatically. We are not mainly just taking task

187
00:10:14,396 --> 00:10:18,556
Speaker 1: for a specific location for a specific customer, say the CIA.

188
00:10:18,676 --> 00:10:21,676
Speaker 1: You know, we are scanning the whole earth, and if

189
00:10:22,156 --> 00:10:25,156
Speaker 1: they want to buy it, that's okay. But other people

190
00:10:25,196 --> 00:10:28,156
Speaker 1: could also buy the same thing, and that's really important.

191
00:10:28,276 --> 00:10:30,396
Speaker 1: Let me ask you about privacy though, is there a

192
00:10:30,396 --> 00:10:33,596
Speaker 1: creepy version of your company? Not that you're that version,

193
00:10:33,636 --> 00:10:35,636
Speaker 1: but like, do you know what I'm saying? Basically, the

194
00:10:35,676 --> 00:10:38,116
Speaker 1: answer is, we don't really get into personal privacy. And

195
00:10:38,276 --> 00:10:42,596
Speaker 1: here's why. With five hundred kilometers up. That's like me

196
00:10:42,796 --> 00:10:48,156
Speaker 1: in San Francisco pointing my camera at Los Angeles and

197
00:10:48,276 --> 00:10:51,556
Speaker 1: expecting to see personal details. It turns out you can't.

198
00:10:51,596 --> 00:10:55,276
Speaker 1: You're too far away that pixel size. Maybe that will

199
00:10:55,356 --> 00:10:58,156
Speaker 1: go in half in time, you know, but you're still

200
00:10:58,196 --> 00:11:01,236
Speaker 1: not going to be identify a person. You know, even

201
00:11:01,276 --> 00:11:03,436
Speaker 1: if it goes in ten x, you still won't be

202
00:11:03,476 --> 00:11:06,796
Speaker 1: able to identify a person. So we're a very long

203
00:11:06,876 --> 00:11:10,556
Speaker 1: way from being able to do anything like that. After

204
00:11:10,596 --> 00:11:13,796
Speaker 1: the break Will's next big problem? How do you turn

205
00:11:13,836 --> 00:11:26,076
Speaker 1: a picture of the earth into useful, actionable information. That's

206
00:11:26,116 --> 00:11:28,116
Speaker 1: the end of the ads. Now we're going back to

207
00:11:28,116 --> 00:11:31,636
Speaker 1: the show. So I want to talk about some of

208
00:11:31,676 --> 00:11:34,876
Speaker 1: the problems you haven't figured out how to solve yet,

209
00:11:34,916 --> 00:11:37,316
Speaker 1: and I think in particular with the data business. Right,

210
00:11:37,316 --> 00:11:39,956
Speaker 1: I mean, it seems like that's really the sort of

211
00:11:39,996 --> 00:11:43,596
Speaker 1: frontier for you. Right. You have these pictures every day

212
00:11:43,596 --> 00:11:45,396
Speaker 1: of the earth, and it's sort of how do you

213
00:11:45,436 --> 00:11:49,676
Speaker 1: make them useful? Right? So what is something? Yeah, what's

214
00:11:49,676 --> 00:11:53,716
Speaker 1: a problem you haven't solved in that part of your business? Well? Many,

215
00:11:54,116 --> 00:11:59,276
Speaker 1: because the use cases are many, And you're right, a picture,

216
00:11:59,716 --> 00:12:02,596
Speaker 1: it can be useful, but most people, most of the

217
00:12:02,636 --> 00:12:05,876
Speaker 1: time want more than a picture. That agriculture of farmer

218
00:12:05,956 --> 00:12:08,196
Speaker 1: actually doesn't want the picture. They might look at it,

219
00:12:08,236 --> 00:12:10,316
Speaker 1: they know what their field look, but they more want

220
00:12:10,436 --> 00:12:12,956
Speaker 1: where do I should I add fertilizers? Where should I order?

221
00:12:14,676 --> 00:12:18,076
Speaker 1: The civil government doesn't want to have pictures. They want

222
00:12:18,076 --> 00:12:20,836
Speaker 1: to know where other roads that are still working after

223
00:12:20,836 --> 00:12:23,476
Speaker 1: the flood or the bridges down after the you know,

224
00:12:23,636 --> 00:12:25,716
Speaker 1: a hurricane or whatever said that they can send in

225
00:12:25,716 --> 00:12:28,876
Speaker 1: the relief supplies. So they want more detailed information that

226
00:12:28,916 --> 00:12:30,636
Speaker 1: it's in principle in those images, but you have to

227
00:12:30,636 --> 00:12:33,516
Speaker 1: extract out. So the big project across all of these

228
00:12:33,556 --> 00:12:37,796
Speaker 1: things that we're doing is something that I call queriable Earth.

229
00:12:37,996 --> 00:12:41,876
Speaker 1: Queriable like searchable, like a search query, searchable earth shot.

230
00:12:42,396 --> 00:12:46,676
Speaker 1: And the reason I think about it the way about this.

231
00:12:46,916 --> 00:12:51,716
Speaker 1: Google indexed what's on the Internet and made it searchable. Right.

232
00:12:51,756 --> 00:12:54,116
Speaker 1: It made a database of where all the pages are,

233
00:12:54,316 --> 00:12:57,716
Speaker 1: what contents are where, so that when you type in whatever,

234
00:12:57,956 --> 00:13:00,676
Speaker 1: you get an answer that's relevant. We have the ability

235
00:13:01,356 --> 00:13:04,516
Speaker 1: with the machine learning. The same tools that pick out

236
00:13:04,716 --> 00:13:07,676
Speaker 1: cats and dogs from pictures that you post online and

237
00:13:07,836 --> 00:13:10,476
Speaker 1: can pull that out is the same too. We can

238
00:13:10,596 --> 00:13:12,756
Speaker 1: use to put out this is a tree, this is

239
00:13:12,796 --> 00:13:15,116
Speaker 1: a road, this is a house, this is a building,

240
00:13:15,196 --> 00:13:18,436
Speaker 1: this is a ship. There's a plane out of our images.

241
00:13:18,916 --> 00:13:22,796
Speaker 1: So we call it object identification using computer vision or

242
00:13:22,836 --> 00:13:27,316
Speaker 1: machine learning or artificial intelligence is the broadest category. Then

243
00:13:27,436 --> 00:13:30,476
Speaker 1: you can imagine that the interface to our data becomes

244
00:13:30,516 --> 00:13:32,916
Speaker 1: less and less the imagery of the earth and more

245
00:13:32,916 --> 00:13:35,956
Speaker 1: and more a little search box and you can put in, hey,

246
00:13:35,996 --> 00:13:39,676
Speaker 1: how many roads are there in India? Give me a

247
00:13:39,716 --> 00:13:43,236
Speaker 1: plot of that, versus time, tell me where the deforestation

248
00:13:43,596 --> 00:13:46,236
Speaker 1: is between this month and last month in the Congo

249
00:13:46,276 --> 00:13:50,276
Speaker 1: basin or the Amazon. The kind of data that you

250
00:13:50,276 --> 00:13:52,116
Speaker 1: should be able to pull out and I call it yeah,

251
00:13:52,156 --> 00:13:56,196
Speaker 1: queriabile Earth. So that's the dream. That's the problem. It's

252
00:13:56,196 --> 00:13:59,236
Speaker 1: a big problem. I mean, it's fundamentally an AI machine

253
00:13:59,276 --> 00:14:01,676
Speaker 1: learning problem. Like, are you guys? Do you guys have

254
00:14:01,716 --> 00:14:04,356
Speaker 1: to solve that AI machine learning problem? Is it sort

255
00:14:04,356 --> 00:14:08,516
Speaker 1: of the frontier of machine learning and really turning images

256
00:14:08,596 --> 00:14:12,956
Speaker 1: into the data right or at least into language? I mean,

257
00:14:12,996 --> 00:14:15,836
Speaker 1: what has to happen? How do you solve that problem? Well,

258
00:14:15,956 --> 00:14:18,316
Speaker 1: luckily computer vision has come a long way and we

259
00:14:18,356 --> 00:14:20,156
Speaker 1: don't have to solve it all ourselves. We are doing

260
00:14:20,236 --> 00:14:23,516
Speaker 1: some of that machine learning ourselves, but we also work

261
00:14:23,556 --> 00:14:27,196
Speaker 1: with partners, especially because a lot of it's bespoke to

262
00:14:27,276 --> 00:14:30,556
Speaker 1: different regions, to different problems, statements. So we have done

263
00:14:30,556 --> 00:14:32,276
Speaker 1: some core things that we think a lot of people

264
00:14:32,316 --> 00:14:36,196
Speaker 1: are interested in, roads, building ships, planes, there's a number

265
00:14:36,236 --> 00:14:38,996
Speaker 1: more who are we going after? And just to be clear,

266
00:14:39,076 --> 00:14:41,276
Speaker 1: can anybody who wants to just go to your website

267
00:14:41,316 --> 00:14:45,436
Speaker 1: and search whatever how many roads are there in Michigan? No, No,

268
00:14:45,596 --> 00:14:49,196
Speaker 1: you can't. Actually you can sign on a trial account

269
00:14:49,196 --> 00:14:51,396
Speaker 1: and you can get some of these capabilities, but generally

270
00:14:51,636 --> 00:14:55,316
Speaker 1: this is set up for at the present more sophisticated users.

271
00:14:55,756 --> 00:14:57,716
Speaker 1: I don't mean to I don't know how sophisticate you

272
00:14:57,756 --> 00:15:00,356
Speaker 1: are personally sitting here in my closet recording a podcast.

273
00:15:00,396 --> 00:15:03,036
Speaker 1: I'm happy to be an unsophisticated US. But you know,

274
00:15:03,316 --> 00:15:07,756
Speaker 1: Google uses our APIs, and those big agg companies use

275
00:15:07,796 --> 00:15:11,236
Speaker 1: our APIs, and they have teams of people and but

276
00:15:11,236 --> 00:15:13,796
Speaker 1: but that's exactly the problem we're trying to crack, is

277
00:15:13,836 --> 00:15:16,756
Speaker 1: to help everyone to be able to have access to

278
00:15:16,796 --> 00:15:19,476
Speaker 1: this data, not just the big organizations that have teams

279
00:15:19,516 --> 00:15:22,556
Speaker 1: of people with PhDs in satellite imagery processing. You know,

280
00:15:22,596 --> 00:15:25,916
Speaker 1: we want to enable it to be that little insurance company,

281
00:15:25,956 --> 00:15:30,716
Speaker 1: that little that little ngo, that university team that's doing

282
00:15:30,756 --> 00:15:33,156
Speaker 1: research on penguins or whatever, you know, Like, we want

283
00:15:33,156 --> 00:15:35,756
Speaker 1: them all to be able to leverage the data. So

284
00:15:35,796 --> 00:15:37,796
Speaker 1: how do we make it easier to use How do

285
00:15:37,876 --> 00:15:40,156
Speaker 1: we help them to build the models that they need

286
00:15:40,356 --> 00:15:43,356
Speaker 1: and build So we're trying to build infrastructure that helps

287
00:15:43,356 --> 00:15:45,636
Speaker 1: them refine this data to make it useful. I mean,

288
00:15:45,676 --> 00:15:48,636
Speaker 1: you need like automation in the middle. Right, there's like

289
00:15:48,676 --> 00:15:50,636
Speaker 1: you got all the pictures, and then you got people

290
00:15:50,636 --> 00:15:52,876
Speaker 1: who want information out of the pictures, and right now

291
00:15:52,916 --> 00:15:56,156
Speaker 1: it's hard and expensive to get the information out of it,

292
00:15:56,716 --> 00:16:14,396
Speaker 1: and so you've got to automate a lot of that process. Bingo. Okay,

293
00:16:14,516 --> 00:16:16,676
Speaker 1: let's get back to the show. We're gonna close with

294
00:16:16,716 --> 00:16:20,276
Speaker 1: the lightning round. Okay, uh, just have a very few

295
00:16:20,396 --> 00:16:25,356
Speaker 1: quick lightning round for the end. What are the chances

296
00:16:25,396 --> 00:16:27,716
Speaker 1: that are Kardashian will go to space? Before you? What

297
00:16:27,876 --> 00:16:30,596
Speaker 1: is a Kardashian? Let me ask it another way, what

298
00:16:30,636 --> 00:16:33,156
Speaker 1: are the chances you'll go to space? I think that

299
00:16:34,396 --> 00:16:36,756
Speaker 1: there's a good chance that anyone that wants to go

300
00:16:36,796 --> 00:16:39,236
Speaker 1: to space will go if they really you know, in

301
00:16:39,276 --> 00:16:42,476
Speaker 1: our lifetime, do you want to go to space? Do

302
00:16:42,516 --> 00:16:44,316
Speaker 1: you want to? I would quite like to, but it's

303
00:16:44,316 --> 00:16:46,116
Speaker 1: not my main goal in life. I would love to

304
00:16:46,156 --> 00:16:47,876
Speaker 1: see the Earth. I'd especially enough to go around the

305
00:16:47,876 --> 00:16:51,436
Speaker 1: moon and look back at the Earth like four times

306
00:16:51,516 --> 00:16:53,636
Speaker 1: wider than the Moon in the sky and be able

307
00:16:53,636 --> 00:16:58,196
Speaker 1: to see the beautiful cradle of civilization. That's a good one.

308
00:16:58,476 --> 00:17:01,836
Speaker 1: Um Star Wars or Star Trek Star check for sure.

309
00:17:02,116 --> 00:17:04,796
Speaker 1: What's one piece of advice you'd give somebody trying to

310
00:17:04,876 --> 00:17:07,956
Speaker 1: solve a hard problem. Um, stick with it, you know.

311
00:17:08,236 --> 00:17:11,556
Speaker 1: I think the biggest thing that differentiates those succeed those

312
00:17:11,596 --> 00:17:13,876
Speaker 1: that don't less the idea than the timing, both of

313
00:17:13,916 --> 00:17:16,916
Speaker 1: which are important, but more of the staying power. It's

314
00:17:17,396 --> 00:17:21,036
Speaker 1: it's perseverance and I think, you know you've got to

315
00:17:21,236 --> 00:17:22,996
Speaker 1: do keep on going in the face of a lot

316
00:17:22,996 --> 00:17:24,916
Speaker 1: of people saying it's not possible. Good. Do you have

317
00:17:24,916 --> 00:17:27,996
Speaker 1: any favorite images that Planet has ever captured. I think

318
00:17:27,996 --> 00:17:31,796
Speaker 1: the very first image we ever captured is an Oregon

319
00:17:31,956 --> 00:17:37,076
Speaker 1: area of partially forested, partially agricultural lands, pasture lands, and

320
00:17:37,836 --> 00:17:40,036
Speaker 1: we could immediately see from that first picture that we

321
00:17:40,076 --> 00:17:43,476
Speaker 1: could identify individual trees in the field. And then we

322
00:17:43,476 --> 00:17:45,116
Speaker 1: were like, oh my god, we can we can do

323
00:17:45,156 --> 00:17:48,516
Speaker 1: this mission. You think you'll ever leave Planet? Leave planet?

324
00:17:48,556 --> 00:17:51,076
Speaker 1: Did you say? Yeah, like the moment when it will

325
00:17:51,116 --> 00:17:53,316
Speaker 1: be time for you to do something else. Last question,

326
00:17:53,356 --> 00:17:57,396
Speaker 1: it's it's definitely my life's work, and I'd be surprised

327
00:17:57,436 --> 00:18:00,516
Speaker 1: if we're not working here in some capacity for a while.

328
00:18:00,556 --> 00:18:04,596
Speaker 1: I mean, look, I enjoy it, and I think I'm

329
00:18:04,636 --> 00:18:12,956
Speaker 1: more likely to leave the planet than Planet. Will Marshall

330
00:18:13,156 --> 00:18:17,396
Speaker 1: is co founder and CEO of Planet, big news in

331
00:18:17,436 --> 00:18:20,236
Speaker 1: the What's Your Problem universe. This week we got our

332
00:18:20,236 --> 00:18:24,876
Speaker 1: own email address, It's problem at pushkin dot Fm. I

333
00:18:24,876 --> 00:18:26,676
Speaker 1: would love to hear what you think of the show.

334
00:18:26,756 --> 00:18:29,596
Speaker 1: If there's topics you want to cover, guests, you want

335
00:18:29,596 --> 00:18:32,436
Speaker 1: to hear, things you like us to do differently, Please

336
00:18:32,556 --> 00:18:36,996
Speaker 1: please do let us know again It's problem at pushkin

337
00:18:37,316 --> 00:18:42,076
Speaker 1: dot Fm. Today's show was edited by Robert Smith, produced

338
00:18:42,076 --> 00:18:46,036
Speaker 1: by Edith Russlo, and engineered by Amanda Kwong. I'm Jacob

339
00:18:46,036 --> 00:18:48,716
Speaker 1: Goldstein and we'll be back next week with another episode

340
00:18:48,756 --> 00:18:55,996
Speaker 1: of What's Your Problem