Recorded at | March 20, 2014 |
---|---|

Event | TED2014 |

Duration (min:sec) | 09:12 |

Video Type | TED Stage Talk |

Words per minute | 217.56 very fast |

Readability (FK) | 58.98 easy |

Speaker | Randall Munroe |

Country | United States of America |

Occupation | programmer, writer, physicist, engineer, geohasher |

Description | American cartoonist, author and engineer |

Official TED page for this talk

**Synopsis**

Web cartoonist Randall Munroe answers simple what-if questions ("what if you hit a baseball moving at the speed of light?") using math, physics, logic and deadpan humor. In this charming talk, a reader's question about Google's data warehouse leads Munroe down a circuitous path to a hilariously over-detailed answer — in which, shhh, you might actually learn something.

1 | 00:12 | So, I have a feature on my website where every week | ||

2 | 00:15 | people submit hypothetical questions | ||

3 | 00:17 | for me to answer, | ||

4 | 00:18 | and I try to answer them using math, science | ||

5 | 00:21 | and comics. | ||

6 | 00:22 | So for example, one person asked, | ||

7 | 00:25 | what would happen if you tried to hit a baseball | ||

8 | 00:26 | pitched at 90 percent of the speed of light? | ||

9 | 00:29 | So I did some calculations. | ||

10 | 00:32 | Now, normally, when an object flies through the air, | ||

11 | 00:34 | the air will flow around the object, | ||

12 | 00:36 | but in this case, the ball would be going so fast | ||

13 | 00:37 | that the air molecules wouldn't have time | ||

14 | 00:39 | to move out of the way. | ||

15 | 00:41 | The ball would smash right into and through them, | ||

16 | 00:44 | and the collisions with these air molecules | ||

17 | 00:45 | would knock away the nitrogen, | ||

18 | 00:47 | carbon and hydrogen from the ball, | ||

19 | 00:49 | fragmenting it off into tiny particles, | ||

20 | 00:51 | and also triggering waves of thermonuclear fusion | ||

21 | 00:53 | in the air around it. | ||

22 | 00:55 | This would result in a flood of x-rays | ||

23 | 00:57 | that would spread out in a bubble | ||

24 | 00:59 | along with exotic particles, | ||

25 | 01:00 | plasma inside, centered on the pitcher's mound, | ||

26 | 01:03 | and that would move away from the pitcher's mound | ||

27 | 01:07 | slightly faster than the ball. | ||

28 | 01:09 | Now at this point, about 30 nanoseconds in, | ||

29 | 01:12 | the home plate is far enough away | ||

30 | 01:14 | that light hasn't had time to reach it, | ||

31 | 01:16 | which means the batter | ||

32 | 01:17 | still sees the pitcher about to throw | ||

33 | 01:20 | and has no idea that anything is wrong. | ||

34 | 01:22 | (Laughter) | ||

35 | 01:24 | Now, after 70 nanoseconds, | ||

36 | 01:27 | the ball will reach home plate, | ||

37 | 01:28 | or at least the cloud of expanding plasma | ||

38 | 01:31 | that used to be the ball, | ||

39 | 01:33 | and it will engulf the bat and the batter | ||

40 | 01:37 | and the plate and the catcher and the umpire | ||

41 | 01:40 | and start disintegrating them all | ||

42 | 01:43 | as it also starts to carry them backward | ||

43 | 01:46 | through the backstop, which also starts to disintegrate. | ||

44 | 01:50 | So if you were watching this whole thing | ||

45 | 01:51 | from a hill, | ||

46 | 01:53 | ideally, far away, | ||

47 | 01:56 | what you'd see is a bright flash of light | ||

48 | 01:57 | that would fade over a few seconds, | ||

49 | 01:59 | followed by a blast wave spreading out, | ||

50 | 02:01 | shredding trees and houses | ||

51 | 02:04 | as it moves away from the stadium, | ||

52 | 02:06 | and then eventually a mushroom cloud | ||

53 | 02:09 | rising up over the ruined city. (Laughter) | ||

54 | 02:12 | So the Major League Baseball rules | ||

55 | 02:13 | are a little bit hazy, | ||

56 | 02:16 | but — (Laughter) — under rule 6.02 and 5.09, | ||

57 | 02:20 | I think that in this situation, | ||

58 | 02:22 | the batter would be considered hit by pitch | ||

59 | 02:25 | and would be eligible to take first base, | ||

60 | 02:27 | if it still existed. | ||

61 | 02:30 | So this is the kind of question I answer, | ||

62 | 02:32 | and I get people writing in with | ||

63 | 02:34 | a lot of other strange questions. | ||

64 | 02:36 | I've had someone write and say, | ||

65 | 02:39 | scientifically speaking, what is the best | ||

66 | 02:41 | and fastest way to hide a body? | ||

67 | 02:43 | Can you do this one soon? | ||

68 | 02:45 | And I had someone write in, | ||

69 | 02:47 | I've had people write in about, | ||

70 | 02:49 | can you prove whether or not you can find love again | ||

71 | 02:51 | after your heart's broken? | ||

72 | 02:53 | And I've had people send in | ||

73 | 02:54 | what are clearly homework questions | ||

74 | 02:56 | they're trying to get me to do for them. | ||

75 | 03:00 | But one week, a couple months ago, | ||

76 | 03:03 | I got a question that was actually about Google. | ||

77 | 03:06 | If all digital data in the world were stored on punch cards, | ||

78 | 03:09 | how big would Google's data warehouse be? | ||

79 | 03:12 | Now, Google's pretty secretive about their operations, | ||

80 | 03:15 | so no one really knows how much data Google has, | ||

81 | 03:18 | and in fact, no one really knows how many data centers Google has, | ||

82 | 03:20 | except people at Google itself. | ||

83 | 03:23 | And I've tried, I've met them a few times, | ||

84 | 03:24 | tried asking them, and they aren't revealing anything. | ||

85 | 03:29 | So I decided to try to figure this out myself. | ||

86 | 03:32 | There are a few things that I looked at here. | ||

87 | 03:34 | I started with money. | ||

88 | 03:36 | Google has to reveal how much they spend, | ||

89 | 03:38 | in general, and that lets you put some caps | ||

90 | 03:40 | on how many data centers could they be building, | ||

91 | 03:44 | because a big data center costs a certain amount of money. | ||

92 | 03:46 | And you can also then put a cap on | ||

93 | 03:49 | how much of the world hard drive market are they taking up, | ||

94 | 03:51 | which turns out, it's pretty sizable. | ||

95 | 03:53 | I read a calculation at one point, | ||

96 | 03:55 | I think Google has a drive failure | ||

97 | 03:56 | about every minute or two, | ||

98 | 04:00 | and they just throw out the hard drive | ||

99 | 04:01 | and swap in a new one. | ||

100 | 04:02 | So they go through a huge number of them. | ||

101 | 04:05 | And so by looking at money, | ||

102 | 04:06 | you can get an idea of how many of these centers they have. | ||

103 | 04:08 | You can also look at power. | ||

104 | 04:10 | You can look at how much electricity they need, | ||

105 | 04:14 | because you need a certain amount of electricity to run the servers, | ||

106 | 04:16 | and Google is more efficient than most, | ||

107 | 04:18 | but they still have some basic requirements, | ||

108 | 04:21 | and that lets you put a limit | ||

109 | 04:23 | on the number of servers that they have. | ||

110 | 04:25 | You can also look at square footage and see | ||

111 | 04:29 | of the data centers that you know, | ||

112 | 04:31 | how big are they? | ||

113 | 04:32 | How much room is that? | ||

114 | 04:33 | How many server racks could you fit in there? | ||

115 | 04:35 | And for some data centers, | ||

116 | 04:37 | you might get two of these pieces of information. | ||

117 | 04:39 | You know how much they spent, | ||

118 | 04:41 | and they also, say, because they had to contract | ||

119 | 04:43 | with the local government | ||

120 | 04:44 | to get the power provided, | ||

121 | 04:46 | you might know what they made a deal to buy, | ||

122 | 04:49 | so you know how much power it takes. | ||

123 | 04:50 | Then you can look at the ratios of those numbers, | ||

124 | 04:53 | and figure out for a data center | ||

125 | 04:54 | where you don't have that information, | ||

126 | 04:56 | you can figure out, | ||

127 | 04:57 | but maybe you only have one of those, | ||

128 | 04:59 | you know the square footage, then you could figure out | ||

129 | 05:01 | well, maybe the power is proportional. | ||

130 | 05:03 | And you can do this same thing with a lot of different quantities, | ||

131 | 05:05 | you know, with guesses about the total amount of storage, | ||

132 | 05:08 | the number of servers, the number of drives per server, | ||

133 | 05:10 | and in each case using what you know | ||

134 | 05:13 | to come up with a model that narrows down | ||

135 | 05:16 | your guesses for the things that you don't know. | ||

136 | 05:18 | It's sort of circling around the number you're trying to get. | ||

137 | 05:20 | And this is a lot of fun. | ||

138 | 05:23 | The math is not all that advanced, | ||

139 | 05:25 | and really it's like nothing more than | ||

140 | 05:28 | solving a sudoku puzzle. | ||

141 | 05:30 | So what I did, I went through all of this information, | ||

142 | 05:35 | spent a day or two researching. | ||

143 | 05:37 | And there are some things I didn't look at. | ||

144 | 05:39 | You could always look at the Google | ||

145 | 05:42 | recruitment messages that they post. | ||

146 | 05:44 | That gives you an idea of where they have people. | ||

147 | 05:46 | Sometimes, when people visit a data center, | ||

148 | 05:47 | they'll take a cell-cam photo and post it, | ||

149 | 05:49 | and they aren't supposed to, | ||

150 | 05:51 | but you can learn things about their hardware that way. | ||

151 | 05:54 | And in fact, you can just look at pizza delivery drivers. | ||

152 | 05:56 | Turns out, they know where all the Google data centers are, | ||

153 | 05:59 | at least the ones that have people in them. | ||

154 | 06:02 | But I came up with my estimate, | ||

155 | 06:04 | which I felt pretty good about, | ||

156 | 06:06 | that was about 10 exabytes of data | ||

157 | 06:09 | across all of Google's operations, | ||

158 | 06:11 | and then another maybe five exabytes or so | ||

159 | 06:15 | of offline storage in tape drives, | ||

160 | 06:17 | which it turns out Google is | ||

161 | 06:18 | about the world's largest consumer of. | ||

162 | 06:21 | So I came up with this estimate, and this is | ||

163 | 06:25 | a staggering amount of data. | ||

164 | 06:26 | It's quite a bit more than any other organization | ||

165 | 06:29 | in the world has, as far as we know. | ||

166 | 06:30 | There's a couple of other contenders, | ||

167 | 06:32 | especially everyone always thinks of the NSA. | ||

168 | 06:35 | But using some of these same methods, | ||

169 | 06:36 | we can look at the NSA's data centers, | ||

170 | 06:38 | and figure out, you know, we don't know what's going on there, | ||

171 | 06:40 | but it's pretty clear that their operation | ||

172 | 06:43 | is not the size of Google's. | ||

173 | 06:44 | Adding all of this up, I came up with | ||

174 | 06:46 | the other thing that we can answer, which is, | ||

175 | 06:48 | how many punch cards would this take? | ||

176 | 06:50 | And so a punch card can hold | ||

177 | 06:53 | about 80 characters, | ||

178 | 06:55 | and you can fit about 2,000 or so cards into a box, | ||

179 | 06:58 | and you put them in, say, | ||

180 | 07:00 | my home region of New England, | ||

181 | 07:02 | it would cover the entire region | ||

182 | 07:04 | up to a depth of a little less than five kilometers, | ||

183 | 07:08 | which is about three times deeper | ||

184 | 07:09 | than the glaciers during the last ice age | ||

185 | 07:11 | about 20,000 years ago. | ||

186 | 07:14 | So this is impractical, but I think | ||

187 | 07:16 | that's about the best answer I could come up with. | ||

188 | 07:19 | And I posted it on my website. I wrote it up. | ||

189 | 07:21 | And I didn't expect to get an answer from Google, | ||

190 | 07:25 | because of course they've been so secretive, | ||

191 | 07:26 | they didn't answer of my questions, | ||

192 | 07:28 | and so I just put it up and said, | ||

193 | 07:29 | well, I guess we'll never know. | ||

194 | 07:31 | But then a little while later | ||

195 | 07:33 | I got a message, a couple weeks later, from Google, | ||

196 | 07:35 | saying, hey, someone here has an envelope for you. | ||

197 | 07:39 | So I go and get it, open it up, | ||

198 | 07:42 | and it's punch cards. (Laughter) | ||

199 | 07:44 | Google-branded punch cards. | ||

200 | 07:47 | And on these punch cards, there are a bunch of holes, | ||

201 | 07:50 | and I said, thank you, thank you, | ||

202 | 07:52 | okay, so what's on here? | ||

203 | 07:53 | So I get some software and start reading it, | ||

204 | 07:55 | and scan them, and it turns out | ||

205 | 07:57 | it's a puzzle. | ||

206 | 07:58 | There's a bunch of code, | ||

207 | 08:00 | and I get some friends to help, | ||

208 | 08:01 | and we crack the code, and then inside that is another code, | ||

209 | 08:04 | and then there are some equations, | ||

210 | 08:05 | and then we solve those equations, | ||

211 | 08:06 | and then finally out pops a message from Google | ||

212 | 08:10 | which is their official answer to my article, | ||

213 | 08:13 | and it said, "No comment." | ||

214 | 08:15 | (Laughter) (Applause) | ||

215 | 08:26 | And I love calculating these kinds of things, | ||

216 | 08:29 | and it's not that I love doing the math. | ||

217 | 08:31 | I do a lot of math, | ||

218 | 08:32 | but I don't really like math for its own sake. | ||

219 | 08:35 | What I love is that it lets you take | ||

220 | 08:37 | some things that you know, | ||

221 | 08:39 | and just by moving symbols around on a piece of paper, | ||

222 | 08:43 | find out something that you didn't know | ||

223 | 08:45 | that's very surprising. | ||

224 | 08:47 | And I have a lot of stupid questions, | ||

225 | 08:49 | and I love that math gives the power | ||

226 | 08:51 | to answer them sometimes. | ||

227 | 08:54 | And sometimes not. | ||

228 | 08:55 | This is a question I got from a reader, | ||

229 | 08:58 | an anonymous reader, | ||

230 | 08:59 | and the subject line just said, "Urgent," | ||

231 | 09:01 | and this was the entire email: | ||

232 | 09:03 | "If people had wheels and could fly, | ||

233 | 09:06 | how would we differentiate them from airplanes?" | ||

234 | 09:09 | Urgent. (Laughter) | ||

235 | 09:11 | And I think there are some questions | ||

236 | 09:15 | that math just cannot answer. | ||

237 | 09:17 | Thank you. | ||

238 | 09:20 | (Applause) |