1
00:00:00,000 --> 00:00:00,212


2
00:00:00,212 --> 00:00:01,420
DAVID J. MALAN: Hello, world!

3
00:00:01,420 --> 00:00:04,900
This is CS50 Live, CS50's
episodic show wherein

4
00:00:04,900 --> 00:00:08,440
we not only look at recent news
in tech, but also explain it.

5
00:00:08,440 --> 00:00:12,820
This week's focus, user
error on a massive scale.

6
00:00:12,820 --> 00:00:15,305
But first, a look back at
our earliest of seasons.

7
00:00:15,305 --> 00:00:21,039


8
00:00:21,039 --> 00:00:21,580
Hello, world!

9
00:00:21,580 --> 00:00:22,120
This is--

10
00:00:22,120 --> 00:00:22,460
ROBOT: CS--

11
00:00:22,460 --> 00:00:23,380
RAMON GALVAN: 50 Live.

12
00:00:23,380 --> 00:00:25,619
I'm Ramon Galvan, filling
in today for David--

13
00:00:25,619 --> 00:00:27,160
DAVID J. MALAN: Who's lost his voice.

14
00:00:27,160 --> 00:00:29,910
RAMON GALVAN: Today he'll be the
Andy Richter to my Conan O'Brien.

15
00:00:29,910 --> 00:00:31,855
[MUSIC PLAYING]

16
00:00:31,855 --> 00:00:54,660


17
00:00:54,660 --> 00:00:57,328
DAVID J. MALAN: And of course,
who could forget Season 2?

18
00:00:57,328 --> 00:00:58,822
[NO SPEECH]

19
00:00:58,822 --> 00:01:07,300


20
00:01:07,300 --> 00:01:10,610
CS50 Live first looks today at GitLab.

21
00:01:10,610 --> 00:01:15,965
GitLab is a popular source code
hosting site, much like GitHub.com,

22
00:01:15,965 --> 00:01:19,460
that developers can use in order
to store their code centrally,

23
00:01:19,460 --> 00:01:21,800
in order to version control
it, share multiple copies,

24
00:01:21,800 --> 00:01:23,660
as well as share it with other users.

25
00:01:23,660 --> 00:01:27,470
Unfortunately, GitLab ran into
a bit of an issue very recently.

26
00:01:27,470 --> 00:01:30,412
The whole incident started
when they saw this.

27
00:01:30,412 --> 00:01:32,120
They support a feature
known as Snippets,

28
00:01:32,120 --> 00:01:34,160
whereby users can
create snippets of code,

29
00:01:34,160 --> 00:01:37,490
much like GitHub Gist, whereby users
can upload small snippets of code

30
00:01:37,490 --> 00:01:39,050
to share them with other people.

31
00:01:39,050 --> 00:01:43,100
Unfortunately, having some
1.5 million snippets of code

32
00:01:43,100 --> 00:01:45,260
created over the course
of just a few days?

33
00:01:45,260 --> 00:01:46,010
Not normal.

34
00:01:46,010 --> 00:01:48,680
In fact, this seemed to be the
result of some spamming behavior

35
00:01:48,680 --> 00:01:50,750
by some adversarial folks online.

36
00:01:50,750 --> 00:01:55,460
Moreover, GitLab also notice that one or
more spammers seemed to be using GitLab

37
00:01:55,460 --> 00:01:59,060
inappropriately, as a content
delivery network, or CDN,

38
00:01:59,060 --> 00:02:02,000
whereby they were serving up
files in ways that they shouldn't.

39
00:02:02,000 --> 00:02:04,580
Now unfortunately,
these kinds of attacks

40
00:02:04,580 --> 00:02:08,180
resulted in a bit of a ripple
effect on their back-end databases.

41
00:02:08,180 --> 00:02:10,437
Particularly, GitLab
posted the following.

42
00:02:10,437 --> 00:02:12,770
"We are experiencing issues
with our production database

43
00:02:12,770 --> 00:02:14,664
and are working to recover."

44
00:02:14,664 --> 00:02:16,580
Now unfortunately, just
minutes later did they

45
00:02:16,580 --> 00:02:19,730
post, "We accidentally
deleted production data

46
00:02:19,730 --> 00:02:22,430
and might have to restore from backup."

47
00:02:22,430 --> 00:02:23,840
Now what exactly happened?

48
00:02:23,840 --> 00:02:27,560
Well, it's quite common for databases
to be replicated from one to another

49
00:02:27,560 --> 00:02:29,430
so that you have a
primary and a secondary,

50
00:02:29,430 --> 00:02:32,990
the latter of which is a backup
of the former in real time.

51
00:02:32,990 --> 00:02:36,140
As part of the diagnosis
challenge for figuring out

52
00:02:36,140 --> 00:02:39,410
why the databases were slowing
down in terms of this replication,

53
00:02:39,410 --> 00:02:42,230
one of GitLab's system
administrators very deliberately

54
00:02:42,230 --> 00:02:44,550
executed a command quite like this.

55
00:02:44,550 --> 00:02:45,640
Now what is this command?

56
00:02:45,640 --> 00:02:47,660
Well at the front of
this command is "sudo,"

57
00:02:47,660 --> 00:02:51,710
which says execute the following
command with administrative, or root,

58
00:02:51,710 --> 00:02:52,760
privileges.

59
00:02:52,760 --> 00:02:54,230
What is the command to be executed?

60
00:02:54,230 --> 00:02:56,270
Well, rm -rf apparently.

61
00:02:56,270 --> 00:02:59,960
And rm, you might know, is to remove
files or folders from a system.

62
00:02:59,960 --> 00:03:02,360
-r though means recursively.

63
00:03:02,360 --> 00:03:05,780
Delete the following thing recursively,
so that any directories inside of that

64
00:03:05,780 --> 00:03:07,190
also get deleted.

65
00:03:07,190 --> 00:03:11,900
And unfortunately, the
"f" -rf means forcibly,

66
00:03:11,900 --> 00:03:15,410
which means don't even prompt
the human to confirm or deny

67
00:03:15,410 --> 00:03:16,910
that he or she wants to do this.

68
00:03:16,910 --> 00:03:20,270
Now the system administrator meant
to execute this command deliberately

69
00:03:20,270 --> 00:03:24,430
on their secondary database,
db2.cluster.gitlab.com,

70
00:03:24,430 --> 00:03:27,320
so that they could resume then
the replication from their primary

71
00:03:27,320 --> 00:03:28,712
to their secondary database.

72
00:03:28,712 --> 00:03:30,920
Unfortunately, it appears
to have been late at night,

73
00:03:30,920 --> 00:03:33,140
and this was a stressful
situation, and darn it

74
00:03:33,140 --> 00:03:38,020
if this command weren't executed
on db1.cluster.gitlab.com,

75
00:03:38,020 --> 00:03:40,130
the actual primary database.

76
00:03:40,130 --> 00:03:42,830
Now no big deal, surely we have
backups all over the place.

77
00:03:42,830 --> 00:03:45,100
So we can just restore from
backup, and our customers

78
00:03:45,100 --> 00:03:47,690
will be perfectly
happy and on their way.

79
00:03:47,690 --> 00:03:52,370
Unfortunately, out of five backup
or replication techniques deployed,

80
00:03:52,370 --> 00:03:57,140
GitLab reported that, "None are working
reliably or set up in the first place."

81
00:03:57,140 --> 00:04:00,650
Indeed, if you'd like to read their
whole post-mortem in which they

82
00:04:00,650 --> 00:04:04,730
discussed exactly what went wrong, and
how, you can check out this URL here.

83
00:04:04,730 --> 00:04:06,680
But the moral of the
story, for our purposes,

84
00:04:06,680 --> 00:04:11,390
is please, please beware
the rm -rf, especially

85
00:04:11,390 --> 00:04:13,730
if you're not just deleting
some directory of your own,

86
00:04:13,730 --> 00:04:16,820
potentially your customers as well.

87
00:04:16,820 --> 00:04:19,550
In other news, you might have
noticed that half of the internet

88
00:04:19,550 --> 00:04:23,300
went down recently, and somehow
this was Amazon.com's fault. Well,

89
00:04:23,300 --> 00:04:26,180
it turns out that Amazon is not
just the e-commerce site that you

90
00:04:26,180 --> 00:04:27,320
might know and use.

91
00:04:27,320 --> 00:04:29,930
They're also one of the world's
largest cloud providers,

92
00:04:29,930 --> 00:04:32,840
where cloud computing is
this technique whereby

93
00:04:32,840 --> 00:04:35,780
other people can run servers,
and have hard drives,

94
00:04:35,780 --> 00:04:38,270
and more services
somewhere in the world.

95
00:04:38,270 --> 00:04:41,210
And you as a customer can
essentially rent those services,

96
00:04:41,210 --> 00:04:44,700
so that your website your application
isn't hosted by you in your own data

97
00:04:44,700 --> 00:04:48,800
center, but in Amazon, or Microsoft,
or Google's own data center.

98
00:04:48,800 --> 00:04:52,940
Now unfortunately, something went wrong
with one of Amazon's cloud services,

99
00:04:52,940 --> 00:04:55,430
something called S3,
simple storage service.

100
00:04:55,430 --> 00:04:59,390
Such that the result, according to
one popular ISP called Level Three,

101
00:04:59,390 --> 00:05:02,870
was outages across
the US, if not beyond,

102
00:05:02,870 --> 00:05:05,870
because these websites pictured--
here is this heap map--

103
00:05:05,870 --> 00:05:08,420
were relying on at least
one of Amazon's services.

104
00:05:08,420 --> 00:05:12,440
In fact you might notice some familiar
names among the websites affected.

105
00:05:12,440 --> 00:05:17,210
Codecademy, Coursera, Docker, Giphy,
GitLab, GitHub, Heroku, Imgur,

106
00:05:17,210 --> 00:05:21,650
Kickstarter, Medium, Quora, Slack,
Travis CI, and many, many more.

107
00:05:21,650 --> 00:05:27,200
In fact, perhaps best was the irony of
a website called, Is It Down Right Now?

108
00:05:27,200 --> 00:05:28,792
being down right now.

109
00:05:28,792 --> 00:05:32,000
This is a website that typically allows
you to check other websites are down,

110
00:05:32,000 --> 00:05:35,300
but if you actually visited that
website during Amazon's outage, would

111
00:05:35,300 --> 00:05:37,320
you have seen an error like this.

112
00:05:37,320 --> 00:05:38,150
Now fair is fair.

113
00:05:38,150 --> 00:05:42,050
Some of CS50's own infrastructure
also went down during this incident,

114
00:05:42,050 --> 00:05:46,140
and that's because CS50 stores not
only some of its largest video files,

115
00:05:46,140 --> 00:05:48,740
but also the data related
to its video player,

116
00:05:48,740 --> 00:05:51,710
on Amazon S3, the cloud
service in question.

117
00:05:51,710 --> 00:05:54,890
So in fact, during that outage, if you
tried to watch one of CS50's videos

118
00:05:54,890 --> 00:05:58,154
in its own player, you probably
would have seen an error screen

119
00:05:58,154 --> 00:05:58,820
quite like this.

120
00:05:58,820 --> 00:06:01,910
Because the video player, which is
JavaScript, or client-side based,

121
00:06:01,910 --> 00:06:06,110
wasn't able to pull the requisite
data from Amazon servers.

122
00:06:06,110 --> 00:06:10,025
Now what is Amazon S3, and what
technically went wrong here?

123
00:06:10,025 --> 00:06:12,400
Well at first glance, it's
all pretty technical sounding.

124
00:06:12,400 --> 00:06:14,860
"Amazon S3 is a simple
key-based object store,"

125
00:06:14,860 --> 00:06:16,592
according to Amazon's documentation.

126
00:06:16,592 --> 00:06:18,550
"Keys can be any string,
and can be constructed

127
00:06:18,550 --> 00:06:20,500
to mimic hierarchical attributes."

128
00:06:20,500 --> 00:06:21,500
But what does that mean?

129
00:06:21,500 --> 00:06:22,708
Well, let's tease this apart.

130
00:06:22,708 --> 00:06:25,480
It's a key-based object store.

131
00:06:25,480 --> 00:06:27,970
Now an object, in this case,
just refers to files, where

132
00:06:27,970 --> 00:06:30,110
a file is just a whole bunch of bits.

133
00:06:30,110 --> 00:06:32,740
But Amazon kind of abstracts
away the notion of a file,

134
00:06:32,740 --> 00:06:35,860
so there isn't really the notion of
files, and folders, and all of that.

135
00:06:35,860 --> 00:06:39,820
There's just objects, which are,
for all intents and purposes, files.

136
00:06:39,820 --> 00:06:43,570
But they are accessible via keys,
which typically are strings,

137
00:06:43,570 --> 00:06:45,790
much like in a hash table, if familiar.

138
00:06:45,790 --> 00:06:48,640
You access some value by
way of some unique key.

139
00:06:48,640 --> 00:06:54,441
So for instance, in CS50, we posted this
first video from fall 2016 at this URL

140
00:06:54,441 --> 00:06:54,940
here.

141
00:06:54,940 --> 00:06:57,040
It's an mp4, which is a video file.

142
00:06:57,040 --> 00:07:00,130
Now it turns out that the
video file actually lives

143
00:07:00,130 --> 00:07:02,830
on a server that similarly
named, but notice what's

144
00:07:02,830 --> 00:07:07,930
in it, cdn.cs50.net.s3.amazonaws.com.

145
00:07:07,930 --> 00:07:12,130
Which is to say that indeed,
within CS50's own CDN--

146
00:07:12,130 --> 00:07:16,352
content delivery network-- the
data itself comes from Amazon.

147
00:07:16,352 --> 00:07:18,310
Now what about the key
that uniquely identifies

148
00:07:18,310 --> 00:07:20,590
our objects, or videos, or other files?

149
00:07:20,590 --> 00:07:25,210
Well this string here with slashes, and
words, and so forth, looks like a file

150
00:07:25,210 --> 00:07:27,130
inside of a bunch of folders, but--

151
00:07:27,130 --> 00:07:29,800
that's fine to think about
it that way-- but it really

152
00:07:29,800 --> 00:07:33,350
is just a unique string
that resembles a file path.

153
00:07:33,350 --> 00:07:36,370
We've adopted a scheme whereby
it looks like these are folders,

154
00:07:36,370 --> 00:07:40,210
simply because it keeps our
data nicely hierarchical.

155
00:07:40,210 --> 00:07:42,400
So what went wrong,
and what did users see?

156
00:07:42,400 --> 00:07:44,680
Well, if you visited
Amazon's status page

157
00:07:44,680 --> 00:07:47,980
on the day in question-- or the
days prior to the days in question--

158
00:07:47,980 --> 00:07:50,290
you would have seen
beautiful green check marks,

159
00:07:50,290 --> 00:07:53,830
from February 27 on back,
whereby all was well.

160
00:07:53,830 --> 00:07:57,580
Green check means good for
the S3 storage service.

161
00:07:57,580 --> 00:08:00,820
Unfortunately, on February 28,
did this thing rear its head.

162
00:08:00,820 --> 00:08:03,850
And suffice it to say, red icon bad.

163
00:08:03,850 --> 00:08:07,954
In fact, in this case it means half of
the internet would appear to be down.

164
00:08:07,954 --> 00:08:10,120
Now, you can read more on
the details of this story,

165
00:08:10,120 --> 00:08:12,790
but let's take a look at
a few of the key moments.

166
00:08:12,790 --> 00:08:17,330
At 2:37 PM Eastern Time on
February 28 did Amazon report this.

167
00:08:17,330 --> 00:08:19,270
"We can confirm high
error rates for requests

168
00:08:19,270 --> 00:08:23,320
made to S3 in the US EAST-1 Region.

169
00:08:23,320 --> 00:08:26,710
We've identified the issue, and are
working to restore normal operations."

170
00:08:26,710 --> 00:08:27,980
Well what does that mean?

171
00:08:27,980 --> 00:08:31,709
Well US EAST-1 Region is simply
one of Amazon's data centers.

172
00:08:31,709 --> 00:08:34,000
Like a lot of cloud providers,
they have data centers--

173
00:08:34,000 --> 00:08:36,760
buildings with lots of servers
and lots of hard drives and more--

174
00:08:36,760 --> 00:08:38,110
all over the world.

175
00:08:38,110 --> 00:08:40,990
And US EAST-1 happens to
be one of the most popular.

176
00:08:40,990 --> 00:08:44,530
It's physically located in northern
Virginia, in the United States.

177
00:08:44,530 --> 00:08:48,430
And because CS50 isn't all that far
away, in Cambridge, Massachusetts,

178
00:08:48,430 --> 00:08:52,810
much of our assets live in
US EAST-1 one by choice.

179
00:08:52,810 --> 00:08:54,190
In fact, it's a trade-off.

180
00:08:54,190 --> 00:08:57,840
We could absolutely replicate our data
across multiple, multiple regions,

181
00:08:57,840 --> 00:09:00,340
and have been much more tolerant
against this kind of fault,

182
00:09:00,340 --> 00:09:02,920
but it's a trade-off between
how much storage you need,

183
00:09:02,920 --> 00:09:06,190
how much money it might cost, and how
much complexity you have to introduce.

184
00:09:06,190 --> 00:09:09,650
So we very consciously put
much of our data in US EAST-1

185
00:09:09,650 --> 00:09:12,260
so that it's as close
to campus as possible.

186
00:09:12,260 --> 00:09:17,152
Now, Amazon explains that the
reason S3 became inaccessible,

187
00:09:17,152 --> 00:09:18,860
and in turn, so many
of these customers--

188
00:09:18,860 --> 00:09:21,670
CS50 among them-- went
offline was as follows.

189
00:09:21,670 --> 00:09:25,120
"The team was debugging an issue
causing the S3 billing system

190
00:09:25,120 --> 00:09:27,100
to progress more slowly than expected."

191
00:09:27,100 --> 00:09:27,830
OK.

192
00:09:27,830 --> 00:09:31,420
"An authorized S3 team member,
using an established playbook,

193
00:09:31,420 --> 00:09:36,130
executed a command which was intended
to remove a small number of servers

194
00:09:36,130 --> 00:09:40,090
for one of the three S3 subsystems
that is used by the building process."

195
00:09:40,090 --> 00:09:41,440
OK.

196
00:09:41,440 --> 00:09:45,840
"Unfortunately, one of the inputs to
the command was entered incorrectly,

197
00:09:45,840 --> 00:09:48,880
and a larger set of servers
was removed than intended."

198
00:09:48,880 --> 00:09:50,950
In other words because of human error.

199
00:09:50,950 --> 00:09:54,820
Literally a typographical error in
the equivalent of a terminal window.

200
00:09:54,820 --> 00:09:58,000
Mistyping a command, did
Amazon take offline--

201
00:09:58,000 --> 00:10:00,790
not just a few servers meant
to diagnose some problem--

202
00:10:00,790 --> 00:10:02,950
but a huge number of servers.

203
00:10:02,950 --> 00:10:06,430
All of which then need to be
rebooted, which takes some time,

204
00:10:06,430 --> 00:10:08,290
and which explains the downtime.

205
00:10:08,290 --> 00:10:11,530
In the real world, this might
be like if Amazon were hungry

206
00:10:11,530 --> 00:10:14,740
for a little bit of
chocolate, and so went over

207
00:10:14,740 --> 00:10:18,850
to the chocolate serving station,
and picked up this here X-ACTO knife,

208
00:10:18,850 --> 00:10:21,670
and just wanted to take a tiny
little piece of the internet offline

209
00:10:21,670 --> 00:10:23,320
so as to enjoy some chocolate.

210
00:10:23,320 --> 00:10:26,280
And so you might just take
off a little corner like this.

211
00:10:26,280 --> 00:10:29,090
Mm-mm, that's a good server.

212
00:10:29,090 --> 00:10:31,420
But that's not what Amazon in fact did.

213
00:10:31,420 --> 00:10:34,960
Amazon, because of a mistyped command,
for which apparently there was not

214
00:10:34,960 --> 00:10:37,600
a sufficient prompting
process to say, "Human are you

215
00:10:37,600 --> 00:10:40,210
sure you want to take
down all these servers?"

216
00:10:40,210 --> 00:10:50,370
Amazon effectively took out this
here saw, turned it on, and bit off

217
00:10:50,370 --> 00:10:53,496
half of the internet.

218
00:10:53,496 --> 00:10:56,032
Mm, that'd good internet.

219
00:10:56,032 --> 00:11:00,040
But, CS50, to be fair, is not
immune to these kind of mistakes.

220
00:11:00,040 --> 00:11:03,270
In fact, here on CS50 Live,
we've made our own fair share.

221
00:11:03,270 --> 00:11:09,540
In fact, why don't we take a look now
at some of CS50 Live's own outtakes.

222
00:11:09,540 --> 00:11:10,320
Hello, world.

223
00:11:10,320 --> 00:11:11,170
This is--

224
00:11:11,170 --> 00:11:11,670
ROBOT: CS--

225
00:11:11,670 --> 00:11:11,990
DAVID J. MALAN: 50--

226
00:11:11,990 --> 00:11:13,040
RAMON GALVAN: Live.

227
00:11:13,040 --> 00:11:15,623
DAVID J. MALAN: So if you see
me trip, if you see me misspeak,

228
00:11:15,623 --> 00:11:18,300
if you screw up, all of that is
happening literally right now,

229
00:11:18,300 --> 00:11:20,205
in Cambridge, Massachusetts.

230
00:11:20,205 --> 00:11:21,600
Oh hi, world.

231
00:11:21,600 --> 00:11:22,540
Drumroll.

232
00:11:22,540 --> 00:11:23,729
Perspect-- or, per second.

233
00:11:23,729 --> 00:11:25,020
Pause the vizio if you'd like--

234
00:11:25,020 --> 00:11:27,340
Book redder.

235
00:11:27,340 --> 00:11:27,990
[INAUDIBLE]

236
00:11:27,990 --> 00:11:32,580
Mark Zuckerberg's favorite patent,
to protect our nuclear missile.

237
00:11:32,580 --> 00:11:34,330
Keeping [INAUDIBLE] as usual.

238
00:11:34,330 --> 00:11:35,250
Bugle itself--

239
00:11:35,250 --> 00:11:36,000
[INAUDIBLE]

240
00:11:36,000 --> 00:11:37,230
Good episode for you.

241
00:11:37,230 --> 00:11:38,280
It's actually quite--

242
00:11:38,280 --> 00:11:39,700
Fine-- and--

243
00:11:39,700 --> 00:11:40,980
Then you may recall--

244
00:11:40,980 --> 00:11:42,720
Ted to this you are here.

245
00:11:42,720 --> 00:11:44,360
TECHNICIAN: This is CS50-- gah.

246
00:11:44,360 --> 00:11:46,360
DAVID J. MALAN: And now
I made the blooper reel.

247
00:11:46,360 --> 00:11:47,130
Fantastic.

248
00:11:47,130 --> 00:11:50,974


249
00:11:50,974 --> 00:11:52,390
SPEAKER 1: Let me read you ending.

250
00:11:52,390 --> 00:11:54,150
That should be the ending.

251
00:11:54,150 --> 00:11:57,560
DAVID J. MALAN: Photos of Jason
Hirschhorn dressed as a pumpkin.

252
00:11:57,560 --> 00:11:58,680
JASON HIRSCHHORN: Oh boy.

253
00:11:58,680 --> 00:12:00,430
I don't know if I want
people to see that.

254
00:12:00,430 --> 00:12:04,020
DAVID J. MALAN: No, now it's photos of
the Jason Hirschhorn dressed as a boy.

255
00:12:04,020 --> 00:12:05,280
SPEAKER 2: And if you're
interested, I can actually

256
00:12:05,280 --> 00:12:06,571
show you how it's going to be--

257
00:12:06,571 --> 00:12:08,250
DAVID J. MALAN: Yeah, absolutely.

258
00:12:08,250 --> 00:12:09,440
SPEAKER 2: He's a little
off balance this morning,

259
00:12:09,440 --> 00:12:10,868
he hasn't had his coffee yet.

260
00:12:10,868 --> 00:12:12,820
[BUS HONKING]

261
00:12:12,820 --> 00:12:13,796
SPEAKER 3: Whoa!

262
00:12:13,796 --> 00:12:14,772
Oh god.

263
00:12:14,772 --> 00:12:19,994


264
00:12:19,994 --> 00:12:20,910
DAVID J. MALAN: From--

265
00:12:20,910 --> 00:12:21,841
where's he from?

266
00:12:21,841 --> 00:12:24,840
SPEAKER 4: Right where my arm is, you
can see like the white characters.

267
00:12:24,840 --> 00:12:26,120
There is the pole.

268
00:12:26,120 --> 00:12:28,410
DAVID J. MALAN: And they don't know
that this is-- you should point here.

269
00:12:28,410 --> 00:12:28,993
SPEAKER 4: Oh.

270
00:12:28,993 --> 00:12:37,155


271
00:12:37,155 --> 00:12:38,780
DAVID J. MALAN: Can you hear me, world?

272
00:12:38,780 --> 00:12:39,905
RAMON GALVAN: Hello, world.

273
00:12:39,905 --> 00:12:41,760
Welcome to CS50 Live, I'm Ramona Galvan.

274
00:12:41,760 --> 00:12:43,090
DAVID J. MALAN: And I--

275
00:12:43,090 --> 00:12:45,560
And I'm David Malan.

276
00:12:45,560 --> 00:12:48,607
RAMON GALVAN: And today I'm
hosting today's episode.

277
00:12:48,607 --> 00:12:49,690
DAVID J. MALAN: But with--

278
00:12:49,690 --> 00:12:52,566
RAMON GALVAN: [BEEP] OK, OK.

279
00:12:52,566 --> 00:12:54,930
TECHNICIAN: Don't say [BEEP] on the air.

280
00:12:54,930 --> 00:12:58,490
RAMON GALVAN: He'll be the Robin to my
Batman, the Andy Richter to my Conan,

281
00:12:58,490 --> 00:12:59,820
the Cheech to my Chong today.

282
00:12:59,820 --> 00:13:02,320
It's most definitely a serious
thing that we're doing today.

283
00:13:02,320 --> 00:13:03,507
This is not a joke.

284
00:13:03,507 --> 00:13:06,340
Dropbox had made quite a fuss lately,
but I know nothing about this.

285
00:13:06,340 --> 00:13:07,310
What is this about?

286
00:13:07,310 --> 00:13:08,650
That was all above me.

287
00:13:08,650 --> 00:13:13,600
And this-- is something I don't know of.

288
00:13:13,600 --> 00:13:15,720
We also took a tour of Third Glass--

289
00:13:15,720 --> 00:13:16,462
Third Deg--

290
00:13:16,462 --> 00:13:17,920
DAVID J. MALAN: [BEEP] right there!

291
00:13:17,920 --> 00:13:22,591
Allows you to swipe credit cards on your
iPhone in order to process payments.

292
00:13:22,591 --> 00:13:24,440
RAMON GALVAN: I have a flip phone.

293
00:13:24,440 --> 00:13:25,315
Let's play the clip.

294
00:13:25,315 --> 00:13:27,870


295
00:13:27,870 --> 00:13:30,650
DAVID J. MALAN: To host the first ever--

296
00:13:30,650 --> 00:13:31,390
RAMON GALVAN: OK.

297
00:13:31,390 --> 00:13:32,560
To host the first ever--

298
00:13:32,560 --> 00:13:35,590


299
00:13:35,590 --> 00:13:37,410
DAVID J. MALAN: I was
in graduate school.

300
00:13:37,410 --> 00:13:39,076
RAMON GALVAN: And I was in fourth grade.

301
00:13:39,076 --> 00:13:41,740
Although I loved Zamyla, I
would much rather not spend--

302
00:13:41,740 --> 00:13:42,760
DAVID J. MALAN: Spend half
as much time with her.

303
00:13:42,760 --> 00:13:43,676
RAMON GALVAN: Exactly.

304
00:13:43,676 --> 00:13:49,870


305
00:13:49,870 --> 00:13:52,000
DAVID J. MALAN: Come on out, Zamyla!

306
00:13:52,000 --> 00:13:53,860
RAMON GALVAN: This was CS50.

307
00:13:53,860 --> 00:13:55,546
DAVID J. MALAN: And this was terrifying.

308
00:13:55,546 --> 00:13:56,920
RAMON GALVAN: This is terrifying.

309
00:13:56,920 --> 00:13:59,590
Made a nice little teaser reel
girl to kind of encapsulate

310
00:13:59,590 --> 00:14:02,110
the debauchery that took place.

311
00:14:02,110 --> 00:14:04,180
DAVID J. MALAN: I love you.

312
00:14:04,180 --> 00:14:07,870
RAMON GALVAN: I love you,
unlike David who circles you.

313
00:14:07,870 --> 00:14:13,250


314
00:14:13,250 --> 00:14:16,420
DAVID J. MALAN: Rumors that-- oh

315
00:14:16,420 --> 00:14:17,860
That's it for CS50 Live.

316
00:14:17,860 --> 00:14:21,610
Thanks so much to CS50 Live's whole
team, Arturo, and Ian, and Christian,

317
00:14:21,610 --> 00:14:25,810
and Doug, and Marina, Marinda, Dan
Coffey, and Cynthia for our brand

318
00:14:25,810 --> 00:14:26,770
new set.

319
00:14:26,770 --> 00:14:29,350
This was CS50.

320
00:14:29,350 --> 00:15:08,572