1
00:00:00,000 --> 00:00:01,792
TOM KEENAN: I'm going to actually start while they try

2
00:00:01,792 --> 00:00:03,250
to set this set up.

3
00:00:03,250 --> 00:00:06,459
I'll tell you a little bit about what we are going to talk about.

4
00:00:06,459 --> 00:00:07,959
The open data is a movement and it has to do a lot

5
00:00:07,959 --> 00:00:10,999
with the idea that the government has information

6
00:00:10,999 --> 00:00:12,292
on you.

7
00:00:12,292 --> 00:00:14,167
We all know a lot about that.

8
00:00:14,167 --> 00:00:18,083
But legitimately they collect all kinds of stuff.

9
00:00:18,083 --> 00:00:19,792
They collect bus schedules.

10
00:00:19,792 --> 00:00:22,125
They collect voting records, things like that.

11
00:00:22,125 --> 00:00:24,375
So the reason there's an open data movement

12
00:00:24,375 --> 00:00:29,125
is because in fact there is a need for it, right?

13
00:00:29,125 --> 00:00:30,751
We want to know the bus time.

14
00:00:30,751 --> 00:00:32,542
And let's face it, governments are not always

15
00:00:32,542 --> 00:00:35,042
the best writers of software.

16
00:00:35,042 --> 00:00:36,167
I'm a professor.

17
00:00:36,167 --> 00:00:38,709
I spent a frantic two hours yesterday trying to deal

18
00:00:38,709 --> 00:00:42,751
with a government grant software in Canada because it was designed

19
00:00:42,751 --> 00:00:44,876
by civil servants.

20
00:00:44,876 --> 00:00:49,834
Often it's best to actually say: Let the community go out and design it.

21
00:00:49,834 --> 00:00:51,999
In New York City, for example, all the subways have

22
00:00:51,999 --> 00:00:56,042
a sign up: "Our transit apps are whiz kid certified."

23
00:00:56,042 --> 00:00:58,334
That means the MTA didn't do them.

24
00:00:58,334 --> 00:01:00,250
They let other companies create applications

25
00:01:00,250 --> 00:01:04,459
to use their data which they make available openly.

26
00:01:04,459 --> 00:01:07,334
One of the reasons governments open up the data is in the interest

27
00:01:07,334 --> 00:01:10,334
of giving better service to customers.

28
00:01:10,334 --> 00:01:12,876
Another one relates to the fact that if they have the data,

29
00:01:12,876 --> 00:01:15,459
they figure sooner other later somebody will put it

30
00:01:15,459 --> 00:01:17,792
out on Wikileaks anyway.

31
00:01:17,792 --> 00:01:19,999
They might as well release what they have.

32
00:01:20,083 --> 00:01:23,250
The ability to go out there and see the government data seems

33
00:01:23,250 --> 00:01:25,667
to be a natural thing.

34
00:01:25,667 --> 00:01:28,751
It resonates with the whole spirit of DEF CON which is data wants

35
00:01:28,751 --> 00:01:30,292
to be free.

36
00:01:30,292 --> 00:01:31,876
Data wants to be open.

37
00:01:31,876 --> 00:01:33,501
As a result, governments are opening up more

38
00:01:33,501 --> 00:01:35,751
and more of their data.

39
00:01:35,751 --> 00:01:37,999
The other thing, of course, is they are cheap.

40
00:01:37,999 --> 00:01:39,459
They don't want to go out there and spend money

41
00:01:39,459 --> 00:01:43,250
on building applications if other people will do it for free.

42
00:01:43,250 --> 00:01:46,125
The reality is, we have a lot of good reasons.

43
00:01:46,125 --> 00:01:48,834
I am not -- this is kind of the big preamble.

44
00:01:48,834 --> 00:01:50,375
I am not against open data.

45
00:01:50,417 --> 00:01:51,918
Open data is great.

46
00:01:51,918 --> 00:01:55,250
There are open data hackathons all over the world.

47
00:01:55,292 --> 00:01:58,834
They have basically the advantage of getting the best minds to take

48
00:01:58,834 --> 00:02:02,667
the government's data and think of creative uses.

49
00:02:02,667 --> 00:02:05,584
If you have four kids and need to take one to the soccer field

50
00:02:05,584 --> 00:02:08,999
and one to the baseball diamond and you want to optimize that,

51
00:02:08,999 --> 00:02:10,834
that's good.

52
00:02:10,876 --> 00:02:13,167
The problem is that a lot of the government the -- data

53
00:02:13,167 --> 00:02:16,626
the government collects is related to people.

54
00:02:16,959 --> 00:02:18,292
I'm still hoping we will see images

55
00:02:18,292 --> 00:02:21,626
because I have images to share with you.

56
00:02:21,626 --> 00:02:24,876
I'll tell you one example relates to voting.

57
00:02:24,876 --> 00:02:27,959
It's pretty obvious in a democracy that we want

58
00:02:27,959 --> 00:02:31,709
to know what the voting results were.

59
00:02:31,709 --> 00:02:33,125
So just, and this is an example that I cooked

60
00:02:33,125 --> 00:02:36,667
up to illustrate one of the problems with open data.

61
00:02:36,667 --> 00:02:38,999
We have a city in Alberta called Edmonton.

62
00:02:38,999 --> 00:02:42,626
It has a wildly popular mayor, Stephen Mandel.

63
00:02:42,626 --> 00:02:46,751
He got over half the votes in every ward.

64
00:02:46,999 --> 00:02:49,876
When they published the votes there were minority

65
00:02:49,876 --> 00:02:53,999
candidates, people who only got a few votes and got zero votes

66
00:02:53,999 --> 00:02:55,959
in some areas.

67
00:02:55,959 --> 00:02:59,999
Say Mr. Dowdele, one of the other candidates, the wife was

68
00:02:59,999 --> 00:03:06,876
in the hospital and she says, "Yes, dear, I voted and I voted for you."

69
00:03:06,876 --> 00:03:09,999
He goes and pulls down the online data and guess what?

70
00:03:09,999 --> 00:03:13,083
He got zero votes in the city wide hospital poll.

71
00:03:13,083 --> 00:03:14,834
That's a type of torturing of data that

72
00:03:14,834 --> 00:03:17,667
the government never anticipated.

73
00:03:17,667 --> 00:03:22,083
They never anticipated we would ask questions like that of the data.

74
00:03:22,083 --> 00:03:24,751
So the term torturing data in my title really relates

75
00:03:24,751 --> 00:03:28,083
to doing what most people in this room love to do, which

76
00:03:28,083 --> 00:03:34,083
is to go out there and find things that you are not supposed to be able to find.

77
00:03:34,083 --> 00:03:36,459
So when people ask what is DEF CON?

78
00:03:36,459 --> 00:03:40,667
I say basically it's not people who are doing things that they shouldn't do.

79
00:03:40,667 --> 00:03:43,667
It's people doing things that they shouldn't be able to do.

80
00:03:43,667 --> 00:03:46,375
And you can do that with data quite effectively.

81
00:03:46,375 --> 00:03:48,959
So again while we are praying to the A/V Gods there,

82
00:03:48,959 --> 00:03:52,209
I'll start telling you about an example.

83
00:03:52,209 --> 00:03:54,999
We'll go through it quickly when the video comes up.

84
00:03:54,999 --> 00:03:55,999
There actually was a challenge called

85
00:03:55,999 --> 00:03:58,999
the open data challenge held in Europe.

86
00:03:59,125 --> 00:04:01,999
They gotten tries from most of the EU countries.

87
00:04:02,375 --> 00:04:05,292
The one that won was from Slovenia.

88
00:04:05,292 --> 00:04:08,501
I can't speak Slovenian, so I can't say the name.

89
00:04:08,501 --> 00:04:12,083
But it translates as: From our taxes.

90
00:04:12,083 --> 00:04:16,292
What they did is they actually took all the government contracts that were

91
00:04:16,292 --> 00:04:18,999
registered online and they started going

92
00:04:18,999 --> 00:04:22,083
out there and putting them together and finding

93
00:04:22,083 --> 00:04:25,083
out what names were in common.

94
00:04:25,083 --> 00:04:26,999
Which people were the directors of different companies,

95
00:04:26,999 --> 00:04:29,209
which people were the lawyers who represented

96
00:04:29,209 --> 00:04:31,250
the deals and so on.

97
00:04:31,250 --> 00:04:35,667
And an interesting thing happened: A woman named Ludmilla decided this

98
00:04:35,667 --> 00:04:38,417
was invading her privacy.

99
00:04:38,501 --> 00:04:42,375
She actually convinced a judge in Slovenia to go out there and

100
00:04:42,375 --> 00:04:46,834
to order the NGO that created this, award winning application,

101
00:04:46,834 --> 00:04:49,292
to take down the data.

102
00:04:49,417 --> 00:04:51,417
Somebody asked me before the talk how technical this

103
00:04:51,417 --> 00:04:53,083
is going to be.

104
00:04:53,083 --> 00:04:55,209
This is about as technical as it gets.

105
00:04:55,209 --> 00:04:56,999
They didn't have that data.

106
00:04:56,999 --> 00:04:59,999
That data was scraped off government databases that were

107
00:04:59,999 --> 00:05:02,876
maintained by the government.

108
00:05:02,876 --> 00:05:04,417
So here we have a judge, and it proves

109
00:05:04,417 --> 00:05:07,250
the old saying that judges are about ten years

110
00:05:07,250 --> 00:05:09,542
behind the average member of society

111
00:05:09,542 --> 00:05:12,459
in understanding of technology.

112
00:05:12,459 --> 00:05:13,459
Yes, that's right.

113
00:05:13,459 --> 00:05:14,459
Yes, thank you.

114
00:05:14,459 --> 00:05:19,292
And so the judge is ordering them to take down this information.

115
00:05:19,292 --> 00:05:22,459
And they can't take it down because they don't really have it.

116
00:05:22,459 --> 00:05:25,459
I suppose they could put filters into their program that somehow allow

117
00:05:25,459 --> 00:05:29,501
them to go out there and to get that information trapped so that it

118
00:05:29,501 --> 00:05:33,501
couldn't be revealed, but it just kind of shows a misunderstanding,

119
00:05:33,501 --> 00:05:36,667
appear fundamental misunderstanding.

120
00:05:36,667 --> 00:05:38,584
Do we have any hopes back there, guys?

121
00:05:39,209 --> 00:05:41,417
(Speaker away from microphone.)    TOM KEENAN:

122
00:05:41,417 --> 00:05:43,834
Hope springs eternal, I guess.

123
00:05:43,834 --> 00:05:45,375
I could hold the screen up and give everybody a telescope

124
00:05:45,375 --> 00:05:47,459
or something like that.

125
00:05:47,709 --> 00:05:50,751
Let's talk about other kinds of open data.

126
00:05:50,751 --> 00:05:52,584
We have a member of the legislative assembly

127
00:05:52,584 --> 00:05:57,209
of Alberta who had the misfortune to do two things wrong.

128
00:05:57,209 --> 00:06:01,584
On a government trip he solicited a prostitute and got caught.

129
00:06:02,751 --> 00:06:04,876
His name is Mike Allen.

130
00:06:04,876 --> 00:06:06,167
It was all over the news.

131
00:06:06,167 --> 00:06:10,083
I said okay, what else can I find out about Mike Allen?

132
00:06:10,083 --> 00:06:11,459
I went to a database.

133
00:06:11,459 --> 00:06:16,083
I went to the Ramsey County, Minnesota, sheriff's office database.

134
00:06:16,125 --> 00:06:19,834
There is poor Mike's information on what he was doing,

135
00:06:19,834 --> 00:06:22,542
the fact that it was gross.

136
00:06:22,542 --> 00:06:25,417
They use the word gross misconduct.

137
00:06:25,417 --> 00:06:27,083
And his home address.

138
00:06:27,083 --> 00:06:29,709
And that's where things get pretty interesting.

139
00:06:29,709 --> 00:06:31,459
If you think about it, having his home address up there,

140
00:06:31,459 --> 00:06:34,417
it's private information in a sense.

141
00:06:34,626 --> 00:06:39,209
Hey, he was arrested, some sheriff in Minneapolis took that and posted it.

142
00:06:39,209 --> 00:06:43,999
We have different countries have different regard for privacy.

143
00:06:43,999 --> 00:06:45,459
So anybody from Germany here?

144
00:06:45,751 --> 00:06:49,083
(One cheer.)    TOM KEENAN: If you're from Germany, you're the best.

145
00:06:49,083 --> 00:06:52,459
The people from Facebook they can't even operate are in Germany.

146
00:06:52,999 --> 00:06:54,876
Everything is against the law there.

147
00:06:56,083 --> 00:06:59,083
Canada is between germ nip and the U.S.

148
00:06:59,083 --> 00:07:01,918
A lot of things you can get away with in the U.S.

149
00:07:02,083 --> 00:07:06,125
in terms of posting information on people would not fly in Canada.

150
00:07:06,125 --> 00:07:09,167
To take that specific example I'm sure if this guy was arrested

151
00:07:09,167 --> 00:07:12,792
they wouldn't willy-nilly put up the information about where

152
00:07:12,792 --> 00:07:16,167
he lives because what is the need for that?

153
00:07:16,167 --> 00:07:18,250
But in the United States there's certainly a tendency to do more

154
00:07:18,250 --> 00:07:19,999
and more of that.

155
00:07:19,999 --> 00:07:22,834
One of the photos if we get to it, I will be able to show you, it's

156
00:07:22,834 --> 00:07:25,292
from Henderson county, Florida.

157
00:07:25,292 --> 00:07:26,999
Henderson County is a place you never want

158
00:07:26,999 --> 00:07:28,751
to get arrested.

159
00:07:29,876 --> 00:07:34,125
Those from Germany who don't know, the sheriffs are elected officials,

160
00:07:34,125 --> 00:07:37,999
dog catchers, people like that, they get to be what they are

161
00:07:37,999 --> 00:07:40,959
by having people vote for them.

162
00:07:40,959 --> 00:07:43,083
The reality is, they go out there and want to show that

163
00:07:43,083 --> 00:07:45,501
they are doing their job.

164
00:07:45,501 --> 00:07:47,083
They are actually a good sheriff.

165
00:07:47,083 --> 00:07:51,918
What better way than to actually arrest people?

166
00:07:51,918 --> 00:07:54,375
And sure enough, in Henderson county, they post the mug shot

167
00:07:54,375 --> 00:07:56,999
of everyone who is arrested.

168
00:07:56,999 --> 00:07:59,083
If you are done for speeding or whatever down there,

169
00:07:59,083 --> 00:08:02,334
and there's a really seedy -- and we may have the slide one

170
00:08:02,334 --> 00:08:04,083
of these days.

171
00:08:04,083 --> 00:08:07,999
If we do, you'll see seedy characters, people you would cross the street

172
00:08:07,999 --> 00:08:11,083
to avoid, who have been arrested.

173
00:08:11,334 --> 00:08:14,709
You'll also see a photo of a 12-year-old boy.

174
00:08:14,834 --> 00:08:17,792
I captured that information off the government database

175
00:08:17,792 --> 00:08:20,751
and used it in my presentation.

176
00:08:20,751 --> 00:08:22,083
Now, I obscured his name.

177
00:08:22,083 --> 00:08:23,417
First name is Bobby.

178
00:08:23,417 --> 00:08:24,667
I obscured the surname.

179
00:08:24,667 --> 00:08:27,999
Put a black bar across his eyes so you can't see

180
00:08:27,999 --> 00:08:33,876
Bobby's personal identity, but the sheriff didn't do that.

181
00:08:33,876 --> 00:08:38,417
The first thing to realize, he's up there, a minor child, 12 years old.

182
00:08:38,876 --> 00:08:42,334
He's being permanently tarred in some ways.

183
00:08:42,334 --> 00:08:44,999
Now, you say, well, not really permanently because after all,

184
00:08:44,999 --> 00:08:47,876
it's only up there for 30 days.

185
00:08:47,876 --> 00:08:52,125
The answer is yeah, but it's been in my presentation for over a year now.

186
00:08:52,125 --> 00:08:56,876
When data is out there, it is -- there's no way to call it back.

187
00:08:56,876 --> 00:09:00,999
There's no way to bring data back into the fold if you've let it loose.

188
00:09:00,999 --> 00:09:03,375
This brings a lot of vulnerability.

189
00:09:03,375 --> 00:09:07,083
The reality is quite a bit of data is being put out there,

190
00:09:07,083 --> 00:09:09,626
captured by people.

191
00:09:09,626 --> 00:09:11,167
They say if you put a photo up on Facebook,

192
00:09:11,167 --> 00:09:15,417
at the very least it's been copied by the NSA, but probably by a lot

193
00:09:15,417 --> 00:09:17,999
of other people as well.

194
00:09:18,250 --> 00:09:24,250
There's no way to call back data from being posted somewhere.

195
00:09:24,876 --> 00:09:26,501
One aspect again, I'm randomly remembering

196
00:09:26,501 --> 00:09:28,334
my presentation.

197
00:09:28,334 --> 00:09:32,250
I usually print out a cheat sheet, but I trusted the tech here.

198
00:09:32,417 --> 00:09:33,417
Ha-ha.

199
00:09:33,876 --> 00:09:36,459
(Cheers and applause and laughter.)    TOM KEENAN:

200
00:09:36,459 --> 00:09:40,083
Another aspect relates to DNA information.

201
00:09:40,083 --> 00:09:42,959
How many of you know ancestry.com?

202
00:09:44,667 --> 00:09:47,999
Or ancestry.co, all the different versions.

203
00:09:48,083 --> 00:09:53,083
I won't ask because it's embarrassing to ask, who has profiles on that?

204
00:09:53,083 --> 00:09:55,542
The business of ancestry, they allow you to sign

205
00:09:55,542 --> 00:10:01,125
up for 14 days for free and do all kinds of exploring with their data.

206
00:10:01,125 --> 00:10:04,792
They are using public information, census data.

207
00:10:05,000 --> 00:10:09,626
They are using military records, prison records, all kinds of things.

208
00:10:09,626 --> 00:10:11,999
Those are particularly useful for Australia.

209
00:10:11,999 --> 00:10:13,792
Lots of prison records down there.

210
00:10:13,792 --> 00:10:15,417
And so they take all this data.

211
00:10:15,417 --> 00:10:18,584
They make it freely available to you for 14 days.

212
00:10:18,584 --> 00:10:23,417
On the 15th day they charge you $299 if you don't cancel the membership.

213
00:10:23,417 --> 00:10:24,999
It's an interesting model.

214
00:10:24,999 --> 00:10:29,042
They take public data and make money from it.

215
00:10:29,042 --> 00:10:31,000
We assume that's probably okay.

216
00:10:31,000 --> 00:10:35,792
The second thing, of course, they, after you leave, after you tell them, hey,

217
00:10:35,792 --> 00:10:39,626
I don't want to sign up for $299, they go out there and

218
00:10:39,626 --> 00:10:41,999
they keep the data.

219
00:10:41,999 --> 00:10:43,834
They know your family tree.

220
00:10:43,834 --> 00:10:46,999
You have enriched their database and guess what?

221
00:10:46,999 --> 00:10:49,209
You can never -- you can cancel your account.

222
00:10:49,209 --> 00:10:51,999
You can check into the Hotel California.

223
00:10:51,999 --> 00:10:53,626
You can never check out your data.

224
00:10:53,626 --> 00:10:55,959
It is permanently part of their database.

225
00:10:55,959 --> 00:10:57,999
A few years ago they got a brain wave.

226
00:10:57,999 --> 00:11:00,334
You know what would make this a lot better?

227
00:11:00,334 --> 00:11:04,459
Send us your DNA and we'll tell you if you are descended from Adam

228
00:11:04,459 --> 00:11:09,459
or Noah or whoever is way, way back in your family tree.

229
00:11:09,626 --> 00:11:13,250
Some people were actually dumb enough to send their DNA.

230
00:11:13,751 --> 00:11:14,959
Doing that, of course, provides

231
00:11:14,959 --> 00:11:17,792
a tremendous amount of information.

232
00:11:17,918 --> 00:11:19,999
If you think about it, with DNA information it's not just

233
00:11:19,999 --> 00:11:21,459
about you.

234
00:11:21,459 --> 00:11:25,999
It's about your siblings, your family, all kinds of people who didn't give any

235
00:11:25,999 --> 00:11:29,626
consent for you to give that information.

236
00:11:29,626 --> 00:11:31,667
So the reality is, putting DNA information

237
00:11:31,667 --> 00:11:33,918
out there is risky.

238
00:11:33,918 --> 00:11:36,083
Giving it up voluntarily is dumb.

239
00:11:36,083 --> 00:11:39,167
And privacy international, the NGO in the U.K.

240
00:11:39,167 --> 00:11:43,083
actually launched a lawsuit against ancestry.com.

241
00:11:43,083 --> 00:11:45,083
Now, I was down in Utah and I thought I would visit

242
00:11:45,083 --> 00:11:48,999
ancestry.com and I tried to find out all about this.

243
00:11:48,999 --> 00:11:52,250
They said oh, well, we just keep the database here.

244
00:11:52,250 --> 00:11:55,667
That's the genetic genealogy project and it's with the smart people

245
00:11:55,667 --> 00:11:57,876
down in California.

246
00:11:57,876 --> 00:12:02,083
So the genetic genealogy project is something to worry about.

247
00:12:02,083 --> 00:12:03,459
The possibility is out there for somebody

248
00:12:03,459 --> 00:12:06,918
to get tremendous amounts of information on you.

249
00:12:06,999 --> 00:12:09,125
Do we have any hope back there?

250
00:12:09,292 --> 00:12:11,375
Hope springs eternal?

251
00:12:12,709 --> 00:12:14,125
(Speaker away from microphone.)    TOM KEENAN:

252
00:12:14,125 --> 00:12:15,999
You're getting closer?

253
00:12:18,999 --> 00:12:21,751
(Speaker away from microphone.)    TOM KEENAN:

254
00:12:21,751 --> 00:12:25,167
Everyone leave your laptop at the door, right?

255
00:12:25,292 --> 00:12:28,999
A couple of principles on dealing with this data.

256
00:12:28,999 --> 00:12:31,918
The New York City released with great fanfare

257
00:12:31,918 --> 00:12:35,292
in 2009 New York City data mine.

258
00:12:35,292 --> 00:12:36,999
They had 210 databases.

259
00:12:36,999 --> 00:12:39,999
There they had interesting things like all the women organizations

260
00:12:39,999 --> 00:12:43,501
in New York City are now in this database.

261
00:12:43,542 --> 00:12:46,999
They realized a day later that they had forgotten to take

262
00:12:46,999 --> 00:12:51,999
out the private e-mail addresses and the secret questions.

263
00:12:52,125 --> 00:12:57,167
So what was your first pet is the most common secret question.

264
00:12:57,167 --> 00:12:59,292
And fluffy is the most common answer.

265
00:12:59,584 --> 00:13:01,999
That information was disclosed out there.

266
00:13:01,999 --> 00:13:04,501
You know, the next day there were only 109 public databases

267
00:13:04,501 --> 00:13:06,999
from New York City because they had to go in there

268
00:13:06,999 --> 00:13:09,999
and take back some of that information.

269
00:13:09,999 --> 00:13:13,375
I want to explain an experiment to you that I did with what

270
00:13:13,375 --> 00:13:16,125
is called open Philly.com.

271
00:13:16,125 --> 00:13:17,626
Philadelphia is one of the leading cities

272
00:13:17,626 --> 00:13:19,626
in making data open.

273
00:13:19,626 --> 00:13:22,417
They went out and actually put online

274
00:13:22,417 --> 00:13:25,417
contribution records.

275
00:13:25,542 --> 00:13:29,375
Because it's an international audience, it varies a lot from country to country,

276
00:13:29,375 --> 00:13:33,999
but in the United States there's laws about campaign contributions.

277
00:13:34,167 --> 00:13:37,083
Those over $200 say that if you are making a contribution,

278
00:13:37,083 --> 00:13:41,999
you have to go occupant and give your name, address and occupation.

279
00:13:42,083 --> 00:13:43,918
Those are public.

280
00:13:45,334 --> 00:13:49,250
Philadelphia took all the contributions, even those of $1.49

281
00:13:49,250 --> 00:13:52,667
for the last seven years retroactively and put them

282
00:13:52,667 --> 00:13:55,375
up in a wonderful database.

283
00:13:55,375 --> 00:13:57,667
We are going to get technical one more time.

284
00:13:57,667 --> 00:14:00,250
It was put up there with a front end that said you can query

285
00:14:00,250 --> 00:14:02,834
this database, but hey, you better not go

286
00:14:02,834 --> 00:14:05,959
out there and actually download it.

287
00:14:05,959 --> 00:14:07,334
It is not for downloading.

288
00:14:07,667 --> 00:14:09,292
Here is how you download it.

289
00:14:09,292 --> 00:14:13,417
Any of my computer science students who didn't get this get an F.

290
00:14:13,417 --> 00:14:17,501
First you say download all the people whose name begins

291
00:14:17,501 --> 00:14:17,999
with A.

292
00:14:17,999 --> 00:14:23,292
Then the last name beginning with A, and then last names, B and C and D.

293
00:14:23,292 --> 00:14:25,501
See the pattern?

294
00:14:25,918 --> 00:14:28,501
Within a few minutes we had a comma separated file

295
00:14:28,501 --> 00:14:31,125
of all these contributions.

296
00:14:31,417 --> 00:14:34,083
Then you torture the data and have fun with it.

297
00:14:34,459 --> 00:14:35,999
What can I find out?

298
00:14:35,999 --> 00:14:37,709
Who in this room knows who Ronald

299
00:14:37,709 --> 00:14:39,250
Rivest is?

300
00:14:43,542 --> 00:14:46,334
There is a candidate, named Shelly something

301
00:14:46,334 --> 00:14:48,999
or other in Philadelphia.

302
00:14:48,999 --> 00:14:50,709
All of her contributions seem to be local

303
00:14:50,709 --> 00:14:53,501
except one came from Massachusetts.

304
00:14:53,501 --> 00:14:54,501
Who was it?

305
00:14:54,501 --> 00:14:55,501
It was Ron Rivest.

306
00:14:55,501 --> 00:14:59,834
He obviously endorsed this candidate enough that he sent money to her.

307
00:15:00,083 --> 00:15:02,918
By plotting the data on a bit of a graph I was actually able

308
00:15:02,918 --> 00:15:05,375
to find some interest things.

309
00:15:05,417 --> 00:15:09,209
The most interesting thing I found was that an awful lot of people

310
00:15:09,209 --> 00:15:12,918
in Philadelphia live in the same place.

311
00:15:12,999 --> 00:15:15,125
1719 Spring Street.

312
00:15:15,375 --> 00:15:17,999
I mean, thousands of people live there.

313
00:15:17,999 --> 00:15:20,876
(Laughter.)    TOM KEENAN: I went: What the heck is this?

314
00:15:20,876 --> 00:15:22,834
I went to Google maps and brought it up.

315
00:15:22,834 --> 00:15:24,751
It is the offices of the International Brotherhood

316
00:15:24,751 --> 00:15:26,999
of Electrical Workers.

317
00:15:27,542 --> 00:15:31,417
So when Guido comes in and is going to get a job,

318
00:15:31,417 --> 00:15:36,709
they say "Come over here and make this contribution."

319
00:15:36,709 --> 00:15:38,375
Oh, yeah, I have to do that.

320
00:15:38,375 --> 00:15:39,501
You fill out the form.

321
00:15:39,501 --> 00:15:41,999
When people do campaign contributions, they put

322
00:15:41,999 --> 00:15:43,999
the address down because they want

323
00:15:43,999 --> 00:15:45,999
the tax receipt.

324
00:15:47,083 --> 00:15:48,999
Maybe he's in the witness protection program

325
00:15:48,999 --> 00:15:52,167
or something and doesn't want to use his address.

326
00:15:52,167 --> 00:15:53,999
They use the address of the union.

327
00:15:53,999 --> 00:15:55,999
A significant number of the contributions

328
00:15:55,999 --> 00:16:01,083
to some candidates come back to one address, the union.

329
00:16:01,083 --> 00:16:06,709
I thought I would run a statistical study.

330
00:16:06,709 --> 00:16:11,542
I found out the most common names in the United States, Jones and Smith.

331
00:16:11,542 --> 00:16:13,375
And I did an analysis.

332
00:16:13,375 --> 00:16:15,918
The only reason really to publish the name and address

333
00:16:15,918 --> 00:16:20,083
of political contributors that I could see is if there's two John Smiths

334
00:16:20,083 --> 00:16:23,250
and you want to know which one it is.

335
00:16:23,250 --> 00:16:25,375
I actually did a statistical test.

336
00:16:25,375 --> 00:16:26,834
I found there were 385 J.

337
00:16:26,834 --> 00:16:33,083
Smiths and the actual power or the need to actually resolve duplicates

338
00:16:33,083 --> 00:16:37,626
came down to only about three or four.

339
00:16:37,626 --> 00:16:41,083
There was almost no information added by having this, but there was

340
00:16:41,083 --> 00:16:43,999
a zipping can't privacy risk.

341
00:16:43,999 --> 00:16:45,709
So what I'm suggesting to governments is that they need

342
00:16:45,709 --> 00:16:48,167
to go out there and think long and hard about how they put

343
00:16:48,167 --> 00:16:50,999
out the data, why they put out the data.

344
00:16:51,375 --> 00:16:53,999
Again, I don't know if we will ever get the images.

345
00:16:53,999 --> 00:16:57,334
We are going to have a flash dance of the images if we get them up here.

346
00:16:57,334 --> 00:17:01,999
You have a complete stalking exercise I did on a notable Albertan.

347
00:17:01,999 --> 00:17:04,334
He was the president of an airline.

348
00:17:04,334 --> 00:17:08,709
I said tell you the truth, I started with the premiere of the province.

349
00:17:08,709 --> 00:17:11,292
I said how well does he protect his privacy?

350
00:17:11,292 --> 00:17:14,125
The first test was, is his phone number unlisted?

351
00:17:14,125 --> 00:17:15,125
It was.

352
00:17:15,125 --> 00:17:16,751
I worked down to other people.

353
00:17:16,751 --> 00:17:19,083
I found the president of a pretty big airline whose home

354
00:17:19,083 --> 00:17:21,083
number is listed.

355
00:17:21,083 --> 00:17:23,792
It is still listed if you look hard enough.

356
00:17:23,959 --> 00:17:24,999
I went from -- (Cheers

357
00:17:24,999 --> 00:17:30,459
and applause.)    TOM KEENAN: Hey, we have the technology!

358
00:17:30,501 --> 00:17:31,876
Okay!

359
00:17:32,083 --> 00:17:33,999
So monkey, or whoever is back.

360
00:17:34,709 --> 00:17:36,709
Work with me darling.

361
00:17:36,709 --> 00:17:37,709
Make it happen.

362
00:17:37,709 --> 00:17:38,834
Go forward.

363
00:17:39,834 --> 00:17:43,083
Just keep -- I'll say next, next, next.

364
00:17:43,584 --> 00:17:44,959
We talked about this.

365
00:17:44,959 --> 00:17:45,959
Stop right there.

366
00:17:45,959 --> 00:17:46,959
New York City.

367
00:17:46,959 --> 00:17:48,459
This is one we didn't talk about.

368
00:17:48,459 --> 00:17:52,459
New York City data mine where the women's organizations were outed.

369
00:17:52,459 --> 00:17:54,083
So there's a problem which is neglecting

370
00:17:54,083 --> 00:17:57,876
to weed and redact data before releasing it.

371
00:17:57,876 --> 00:17:58,876
Next slide.

372
00:17:58,918 --> 00:18:00,626
Yes, Toronto.

373
00:18:00,626 --> 00:18:03,250
You call 311, you complain about something.

374
00:18:03,250 --> 00:18:06,626
Sure enough, in Toronto they built a database of that.

375
00:18:06,626 --> 00:18:09,792
They are supposed to be careful to anonymize it.

376
00:18:09,792 --> 00:18:11,959
We have six digit postal codes.

377
00:18:11,959 --> 00:18:13,999
They are supposed to put in the first three.

378
00:18:13,999 --> 00:18:15,292
I looked at the database.

379
00:18:15,292 --> 00:18:18,999
Sometimes they go and put actual intersections.

380
00:18:19,834 --> 00:18:23,417
Sloppy, lazy data entry, the next problem.

381
00:18:23,417 --> 00:18:24,417
Next slide.

382
00:18:24,501 --> 00:18:27,584
This is actually from my place in Calgary.

383
00:18:27,584 --> 00:18:31,292
Do any of you know C click fix, you want to report a pothole or

384
00:18:31,292 --> 00:18:35,125
a place where criminals hang out, you go anonymously

385
00:18:35,125 --> 00:18:38,999
on this system and you enter the data.

386
00:18:38,999 --> 00:18:40,584
Sure enough, in Calgary somebody hates

387
00:18:40,584 --> 00:18:43,334
the car wash in his neighborhood.

388
00:18:43,334 --> 00:18:47,083
Every day he puts in excessive noise, dangerous ice.

389
00:18:47,083 --> 00:18:49,501
He has a hate on for that car wash.

390
00:18:49,501 --> 00:18:50,501
Next slide.

391
00:18:50,959 --> 00:18:53,334
Okay, this is the European data channel.

392
00:18:53,334 --> 00:18:54,334
Next slide.

393
00:18:54,459 --> 00:18:55,999
Who won?

394
00:18:55,999 --> 00:18:58,334
There it is in Slovenia, from our taxes.

395
00:18:58,334 --> 00:18:59,334
Next slide.

396
00:18:59,709 --> 00:19:01,709
Okay, who wouldn't agree with that?

397
00:19:01,709 --> 00:19:04,167
Well, the judge ordered them to take down the data.

398
00:19:04,167 --> 00:19:05,167
Next slide.

399
00:19:05,375 --> 00:19:07,083
A little technical side.

400
00:19:07,083 --> 00:19:08,834
They couldn't take down the data.

401
00:19:08,834 --> 00:19:11,167
Judges lag behind technology much we know that.

402
00:19:11,167 --> 00:19:13,999
Next slide, election results.

403
00:19:13,999 --> 00:19:15,876
There's the guy who got all the votes.

404
00:19:15,876 --> 00:19:16,876
Next slide.

405
00:19:16,876 --> 00:19:17,999
Did my wife vote for me?

406
00:19:17,999 --> 00:19:19,209
Apparently not because all these people got

407
00:19:19,209 --> 00:19:20,959
zero slides.

408
00:19:21,999 --> 00:19:24,167
Philadelphia, finance records.

409
00:19:24,167 --> 00:19:27,709
Next slide, there it is, full home addresses provided.

410
00:19:27,709 --> 00:19:28,709
Next slide.

411
00:19:28,792 --> 00:19:31,999
Okay, that's the kind of computer that we had.

412
00:19:31,999 --> 00:19:33,459
I had one of those come doors.

413
00:19:33,459 --> 00:19:34,459
Anybody have a pet?

414
00:19:34,459 --> 00:19:35,459
Yeah.

415
00:19:35,459 --> 00:19:37,709
1970s when they wrote the election law.

416
00:19:37,709 --> 00:19:39,375
Maybe they're a little out of date.

417
00:19:39,375 --> 00:19:40,375
Next slide.

418
00:19:40,501 --> 00:19:42,334
There's the kind of stuff I can get.

419
00:19:42,334 --> 00:19:45,375
Ronald Rivest's home address if you want to write to him.

420
00:19:47,999 --> 00:19:50,999
Yes, that's Ron and all of his family gave

421
00:19:50,999 --> 00:19:54,125
to Barack Obama in this case.

422
00:19:54,125 --> 00:19:55,125
Next slide.

423
00:19:55,292 --> 00:19:57,083
Let's keep going.

424
00:19:57,083 --> 00:19:58,083
Next slide.

425
00:19:58,083 --> 00:20:01,083
So the files were downloadable.

426
00:20:01,083 --> 00:20:02,083
Next slide.

427
00:20:02,083 --> 00:20:05,501
Common names, Smith, Johnson, Williams.

428
00:20:05,501 --> 00:20:06,999
That's the number of Smiths and Johnsons,

429
00:20:06,999 --> 00:20:09,626
very small number of duplicates.

430
00:20:09,626 --> 00:20:10,626
Next slide.

431
00:20:10,918 --> 00:20:14,999
It was really useless and when I looked at the address on spring street,

432
00:20:14,999 --> 00:20:16,709
next slide.

433
00:20:16,709 --> 00:20:19,999
There it is, International Brotherhood of Electrical Workers.

434
00:20:24,125 --> 00:20:27,292
Inconsistent data like some people giving home address,

435
00:20:27,292 --> 00:20:29,626
some giving the union.

436
00:20:29,626 --> 00:20:32,999
And the why are we requiring an address needs to be examined

437
00:20:32,999 --> 00:20:35,709
in the light of new tools.

438
00:20:35,709 --> 00:20:36,709
Next slide.

439
00:20:36,709 --> 00:20:38,918
There's the guy, Clyde Beddow.

440
00:20:39,876 --> 00:20:42,292
There's his home address from the phone book.

441
00:20:42,292 --> 00:20:43,918
Took out the phone number.

442
00:20:43,918 --> 00:20:44,918
It's still valid.

443
00:20:45,000 --> 00:20:47,999
Property tax assessment, next slide.

444
00:20:48,167 --> 00:20:50,876
Details of his property tax.

445
00:20:50,876 --> 00:20:51,876
Next slide.

446
00:20:51,876 --> 00:20:53,250
There he is plotted on a map.

447
00:20:53,250 --> 00:20:56,292
His house is worth 710 less.

448
00:20:56,292 --> 00:20:57,501
The neighbors less.

449
00:20:57,501 --> 00:21:01,375
This is an industry of writing letters.

450
00:21:01,417 --> 00:21:05,751
Dear Mr. Beddow, you are assessed this much money.

451
00:21:05,751 --> 00:21:09,083
Do you know that your neighbor is assessed at 414,000?

452
00:21:09,083 --> 00:21:11,751
Would you like us to appeal your taxes?

453
00:21:11,792 --> 00:21:14,375
The city had to shut down the database.

454
00:21:14,999 --> 00:21:19,375
They now have technical and legal safeguards which are dumb.

455
00:21:19,501 --> 00:21:23,083
You can only query ten times from an IP address in a day.

456
00:21:23,083 --> 00:21:25,751
Anybody in this room can spoof that one.

457
00:21:25,751 --> 00:21:26,751
Next slide.

458
00:21:27,083 --> 00:21:28,501
Next one?

459
00:21:28,959 --> 00:21:31,999
Here is Hernando County, the sheriff, next slide.

460
00:21:31,999 --> 00:21:32,999
Smiley face.

461
00:21:33,417 --> 00:21:36,626
There is not so friendly people who got arrested for things

462
00:21:36,626 --> 00:21:40,542
like uttering forged instrument and contempt of court.

463
00:21:40,542 --> 00:21:41,542
Next slide.

464
00:21:41,667 --> 00:21:43,709
Poor Bobby, that's real.

465
00:21:43,959 --> 00:21:48,999
1997, that made him 12 years, 12 months at the time of arrest.

466
00:21:48,999 --> 00:21:50,375
Shame on you, sheriff.

467
00:21:50,999 --> 00:21:52,959
Data journalism.

468
00:21:52,959 --> 00:21:54,918
I lived in the Bronx for a long time.

469
00:21:54,918 --> 00:22:00,250
The Backman County newspaper listed people with pistol permits.

470
00:22:00,250 --> 00:22:03,751
Sure enough, one of them lived in the Bronx on Milltown Road

471
00:22:03,751 --> 00:22:06,959
a few blocks from where I lived.

472
00:22:07,209 --> 00:22:09,125
This is public information.

473
00:22:09,125 --> 00:22:10,999
You get interesting results.

474
00:22:10,999 --> 00:22:12,334
People have pistol, multiple pistol permits

475
00:22:12,334 --> 00:22:15,542
and live opposite an elementary school.

476
00:22:15,542 --> 00:22:16,542
Next slide.

477
00:22:19,417 --> 00:22:22,083
Should the rich have less privacy?

478
00:22:22,083 --> 00:22:23,292
When we go out and do data journalism

479
00:22:23,292 --> 00:22:26,083
on rich people like corporate directors we can go

480
00:22:26,083 --> 00:22:30,876
out there and find people ab abusing the SNAP food stamp program.

481
00:22:30,918 --> 00:22:35,459
Interesting societal questions, do we go after those people or just

482
00:22:35,459 --> 00:22:39,292
the easy pickings, the high profile ones?

483
00:22:39,709 --> 00:22:41,542
Indirect risk.

484
00:22:41,542 --> 00:22:43,999
People collecting this data like check point.

485
00:22:44,542 --> 00:22:47,584
Ancestry.com, they have 6 billion records mostly

486
00:22:47,584 --> 00:22:49,999
courtesy of the public.

487
00:22:49,999 --> 00:22:50,999
Next slide.

488
00:22:50,999 --> 00:22:52,792
There is your DNA test.

489
00:22:52,792 --> 00:22:56,501
They want your DNA and charge you $99 to take it from you.

490
00:22:56,501 --> 00:22:58,918
Next slide or if you're in New York City there's mobile

491
00:22:58,918 --> 00:23:00,584
DNA testing.

492
00:23:00,584 --> 00:23:02,999
All you have to do is flag him down, there he is.

493
00:23:02,999 --> 00:23:05,999
He will tell you who your daddy was right on the spot.

494
00:23:05,999 --> 00:23:07,083
Or who your daddy isn't.

495
00:23:07,083 --> 00:23:10,584
(Cheers and applause.)    TOM KEENAN: Next

496
00:23:10,584 --> 00:23:12,999
slide, please.

497
00:23:12,999 --> 00:23:16,334
There is the complaint against him from privacy international.

498
00:23:16,334 --> 00:23:17,334
Next slide.

499
00:23:17,334 --> 00:23:19,375
Tools that we use for this.

500
00:23:19,375 --> 00:23:21,959
Inquiry mind, that's the important one.

501
00:23:21,959 --> 00:23:25,999
Having a motive like financial motive, ID theft, et cetera.

502
00:23:25,999 --> 00:23:31,501
Scripting language, Scraper Wiki again, getting Python, PHP scripts.

503
00:23:31,501 --> 00:23:32,999
All this stuff is out there.

504
00:23:32,999 --> 00:23:36,709
There are libraries of scripts, if you want to do it, go to a hackathon.

505
00:23:36,999 --> 00:23:38,999
Some friendly person will tell you how.

506
00:23:46,083 --> 00:23:48,876
Companies like ChoicePoint actually go out there and

507
00:23:48,876 --> 00:23:51,459
they pay high school students to go into the basement

508
00:23:51,459 --> 00:23:54,292
of courthouses and hand copy people's divorce settlements

509
00:23:54,292 --> 00:23:57,375
because you are not allowed to photocopy them and not allowed

510
00:23:57,375 --> 00:24:01,709
to download them electronically, but they are public information.

511
00:24:01,709 --> 00:24:04,501
Once that stuff gets put into a public file as it

512
00:24:04,501 --> 00:24:08,834
is at ChoicePoint it can be accessed for a fee.

513
00:24:08,834 --> 00:24:10,918
One guy can never get a job.

514
00:24:10,918 --> 00:24:11,918
He didn't know why.

515
00:24:11,918 --> 00:24:12,999
He was an engineer.

516
00:24:12,999 --> 00:24:14,999
He would get interviewed, never hire.

517
00:24:14,999 --> 00:24:19,334
He finally got a friend to pull his private file from choice point.

518
00:24:19,334 --> 00:24:20,334
What was it?

519
00:24:20,334 --> 00:24:21,999
He was a convicted murderer.

520
00:24:21,999 --> 00:24:24,375
Well, he wasn't actually a convicted murderer!

521
00:24:24,375 --> 00:24:26,667
They got his Social Security number wrong.

522
00:24:26,709 --> 00:24:29,334
His friend said you know why you're not getting a job.

523
00:24:29,334 --> 00:24:30,876
They said you're a murderer.

524
00:24:30,876 --> 00:24:31,876
He said no, I'm not.

525
00:24:31,876 --> 00:24:32,876
Next slide.

526
00:24:32,876 --> 00:24:34,083
Here are the dirty dozen.

527
00:24:34,709 --> 00:24:37,751
Sloppy data entry, malicious information, lag

528
00:24:37,751 --> 00:24:40,459
between law and policy, making inferences

529
00:24:40,459 --> 00:24:45,083
from the small numbers like very few people who voted.

530
00:24:45,083 --> 00:24:47,501
Assuming nobody like me will torture the data.

531
00:24:47,501 --> 00:24:49,125
My name is Tom and I torture data.

532
00:24:49,125 --> 00:24:50,125
Next.

533
00:24:50,709 --> 00:24:52,167
Inconsistent data.

534
00:24:52,167 --> 00:24:55,375
Where, you know, I did a project with the Rand corporation years ago

535
00:24:55,375 --> 00:24:59,999
where we were collecting data on New York City fire trucks.

536
00:24:59,999 --> 00:25:03,584
We had people getting from the battery, like battery park to Harlem

537
00:25:03,584 --> 00:25:07,459
in a fire truck in three minutes and 402_nd_s.

538
00:25:07,501 --> 00:25:09,709
We couldn't figure out how that could be.

539
00:25:09,709 --> 00:25:10,999
They are not helicopters.

540
00:25:11,250 --> 00:25:12,999
We watched them.

541
00:25:12,999 --> 00:25:13,709
And what happened, the guy who sits

542
00:25:13,709 --> 00:25:16,459
next to the driver was recording the data.

543
00:25:16,584 --> 00:25:19,375
They would go out and put fires out all day and say oh, we forgot

544
00:25:19,375 --> 00:25:22,999
to do that, but the Rand guys, we need to fix it.

545
00:25:22,999 --> 00:25:24,167
They made up the data.

546
00:25:24,834 --> 00:25:29,292
Jigsawing, taking data public and private and putting it together.

547
00:25:29,292 --> 00:25:32,125
Facial recognition, big area coming up.

548
00:25:32,125 --> 00:25:33,667
Your face can be.

549
00:25:35,083 --> 00:25:39,709
Somebody stood here last year and said he can take your photo

550
00:25:39,709 --> 00:25:43,918
on Facebook, compare it to a photo on Match.com, and tell

551
00:25:43,918 --> 00:25:48,083
if you're Sexy Babe or you're Hung Dude 204.

552
00:25:48,501 --> 00:25:52,501
He can disambiguate those using your face.

553
00:25:53,999 --> 00:25:57,375
Recommendations, we have to scan the file for PII things that need

554
00:25:57,375 --> 00:25:59,083
to be included.

555
00:26:00,918 --> 00:26:02,999
This is on the CD.

556
00:26:02,999 --> 00:26:04,417
I don't need to read it to you.

557
00:26:04,417 --> 00:26:05,417
Next slide.

558
00:26:06,083 --> 00:26:08,918
I want to tell you a little bit about this image.

559
00:26:08,999 --> 00:26:12,876
There's a great project out there, concepts for all.

560
00:26:12,876 --> 00:26:16,292
It basically is, I want to end on a hopeful note.

561
00:26:16,292 --> 00:26:19,918
There was a poster program at the University of Quebec in Montreal,

562
00:26:19,918 --> 00:26:21,918
a design program going to be shut

563
00:26:21,918 --> 00:26:25,834
down because it's so expensive to print posters.

564
00:26:26,083 --> 00:26:29,167
My friend discovered the Internet.

565
00:26:29,167 --> 00:26:31,999
Now all these images get posted on the Internet.

566
00:26:31,999 --> 00:26:33,709
They are beautiful images.

567
00:26:33,709 --> 00:26:36,999
The only condition for using them for something like this

568
00:26:36,999 --> 00:26:41,250
is that you say this was courtesy of Elyse Pachef.

569
00:26:42,918 --> 00:26:44,626
I did that.

570
00:26:44,999 --> 00:26:47,417
I recommend that you talk to the government.

571
00:26:47,417 --> 00:26:48,626
We are the people.

572
00:26:48,626 --> 00:26:49,751
We are the democracy.

573
00:26:49,751 --> 00:26:53,999
We get to decide supposedly what our governments do to or for us.

574
00:26:53,999 --> 00:26:56,792
This is an area you will be hearing much more about.

575
00:26:56,792 --> 00:26:58,417
I promised to stay on time.

576
00:26:58,417 --> 00:27:00,542
I think I have five minutes for questions.

577
00:27:00,542 --> 00:27:01,542
Is that about right?

578
00:27:01,876 --> 00:27:02,876
Okay.

579
00:27:02,876 --> 00:27:04,501
So let's have the first question.

580
00:27:04,501 --> 00:27:05,501
Thank you!

581
00:27:09,876 --> 00:27:11,250
(Cheers and applause.)    TOM KEENAN:

582
00:27:11,250 --> 00:27:14,999
Thanks for all the hard working goons who made this come up there.

583
00:27:14,999 --> 00:27:17,209
We paved the way for first speakers.

584
00:27:17,209 --> 00:27:18,999
It's hard to get the first question.

585
00:27:18,999 --> 00:27:20,792
I'll answer the second question.

586
00:27:23,083 --> 00:27:26,959
Who has a question?

587
00:27:26,959 --> 00:27:27,959
Okay.

588
00:27:27,959 --> 00:27:28,959
Turn him on, please.

589
00:27:34,999 --> 00:27:37,918
AUDIENCE: I would rather stay off the stage.

590
00:27:37,918 --> 00:27:39,292
I don't want to be recognized.

591
00:27:39,667 --> 00:27:42,417
There is a movement in the United States to crack

592
00:27:42,417 --> 00:27:46,250
down on perceived voter registration fraud.

593
00:27:46,250 --> 00:27:50,083
It's mostly bullshit, but that's the way to keep black people from voting.

594
00:27:50,501 --> 00:27:53,999
(Cheers.)    AUDIENCE: But in general, I think governments feel

595
00:27:53,999 --> 00:27:58,501
like if you require people to provide an address it keeps them honest even

596
00:27:58,501 --> 00:28:01,501
if it isn't particularly useful.

597
00:28:01,501 --> 00:28:05,334
Is there another way you can think of to keep people more honest.

598
00:28:05,334 --> 00:28:07,083
TOM KEENAN: Is there another way?

599
00:28:07,083 --> 00:28:08,999
AUDIENCE: Should we stop worrying about it,

600
00:28:08,999 --> 00:28:12,999
actual voter fraud where somebody pretends to vote or votes twice,

601
00:28:12,999 --> 00:28:15,999
it's hard to find an instance.

602
00:28:15,999 --> 00:28:16,999
What do you think?

603
00:28:16,999 --> 00:28:20,292
TOM KEENAN: You can take their DNA, all kinds of things you can do.

604
00:28:20,501 --> 00:28:21,999
The point of the address.

605
00:28:21,999 --> 00:28:24,542
AUDIENCE: (Speaker away from microphone.)    TOM KEENAN:

606
00:28:24,542 --> 00:28:27,250
You can do -- he asked how you can do something

607
00:28:27,250 --> 00:28:30,999
other than addresses to eliminate voter fraud and I joked you

608
00:28:30,999 --> 00:28:33,083
can take people's DNA.

609
00:28:33,083 --> 00:28:36,250
There are countries in the world where people vote.

610
00:28:36,751 --> 00:28:39,209
They get their finger marked with ink and sometimes

611
00:28:39,209 --> 00:28:42,999
they don't want to vote because they might get that finger cut

612
00:28:42,999 --> 00:28:46,459
off by somebody or be punished by voting.

613
00:28:46,626 --> 00:28:51,542
There are alternative ways of doing it, but I can't think of too many other ways.

614
00:28:51,542 --> 00:28:53,999
If the contribution is small enough, the reality is giving

615
00:28:53,999 --> 00:28:56,292
the address is optional.

616
00:28:56,292 --> 00:28:59,334
Give the address of the union and never go pick up the result.

617
00:29:00,083 --> 00:29:04,999
Campaign laws, the big idea is that the campaign law needs to be reviewed.

618
00:29:04,999 --> 00:29:07,083
It was passed in the 1970s.

619
00:29:07,959 --> 00:29:10,459
We didn't have the technology to do the kinds of things we do now

620
00:29:10,459 --> 00:29:11,999
with the data.

621
00:29:11,999 --> 00:29:13,918
It's only going to get bigger.

622
00:29:13,918 --> 00:29:16,751
The data now is going to stick around forever.

623
00:29:16,751 --> 00:29:20,667
So they will go back seven years, but maybe 27 years.

624
00:29:20,751 --> 00:29:21,999
Yes?

625
00:29:21,999 --> 00:29:25,999
AUDIENCE: (Speaker away from microphone.)    TOM KEENAN:

626
00:29:25,999 --> 00:29:30,083
Scraper Wiki has a whole bunch of tools.

627
00:29:31,501 --> 00:29:35,501
I used Excel because that's all I really had to use to take that,

628
00:29:35,501 --> 00:29:40,999
but in the presentation there are those lists of Scraper Wiki and so on.

629
00:29:40,999 --> 00:29:42,876
There are a bunch of places where you have customized

630
00:29:42,876 --> 00:29:47,375
scripts that allow you to go out there and scrape public databases.

631
00:29:47,501 --> 00:29:49,501
The world is your oyster.

632
00:29:49,501 --> 00:29:54,083
I think every municipality now has something like that.

633
00:29:54,125 --> 00:29:56,501
We find more and more.

634
00:29:56,501 --> 00:29:59,999
I haven't found one I don't have problems with.

635
00:29:59,999 --> 00:30:03,000
The challenge between now and next year is to find more holes

636
00:30:03,000 --> 00:30:07,042
in open government and bring them to our attention.

637
00:30:08,250 --> 00:30:10,167
One more question maybe?

638
00:30:12,584 --> 00:30:14,042
Going once?

639
00:30:14,584 --> 00:30:17,334
You have one more.

640
00:30:17,334 --> 00:30:19,334
TOM KEENAN: Good.

641
00:30:19,334 --> 00:30:21,999
AUDIENCE: Do you know if any of those terms of service

642
00:30:21,999 --> 00:30:25,792
for public data requests have been tested in courts?

643
00:30:25,792 --> 00:30:28,876
TOM KEENAN: I don't, and I actually asked Marcia Hoffmann

644
00:30:28,876 --> 00:30:32,375
that question and she's researching that.

645
00:30:32,375 --> 00:30:34,999
I thought that's an interesting thing.

646
00:30:34,999 --> 00:30:38,501
All I can tell you in Calgary, I'll summarize the story.

647
00:30:38,501 --> 00:30:42,000
Because of people doing what I did and also because of that secondary use

648
00:30:42,000 --> 00:30:45,834
of the data to actually try to commercialize it, the city shut

649
00:30:45,834 --> 00:30:49,709
the database down for a period of time and when it came back there

650
00:30:49,709 --> 00:30:52,709
was a very elaborate terms of use.

651
00:30:52,709 --> 00:30:54,584
I violated it here today because it says this is only

652
00:30:54,584 --> 00:30:57,501
for checking your tax records because when it first came

653
00:30:57,501 --> 00:31:01,375
out people were checking their boss's house assessment.

654
00:31:01,375 --> 00:31:04,501
They were checking their ex-wife, everybody that they knew.

655
00:31:04,876 --> 00:31:06,459
They now control it.

656
00:31:06,459 --> 00:31:08,959
So what they've done, and it's a good point.

657
00:31:08,959 --> 00:31:10,999
They have two different levels of access.

658
00:31:10,999 --> 00:31:13,334
If you want to just know -- I should tell you, you are allowed

659
00:31:13,334 --> 00:31:15,999
to know what your neighbor is assessed because that

660
00:31:15,999 --> 00:31:18,083
is an issue of fairness.

661
00:31:18,083 --> 00:31:20,999
If you are assessed on fair, you have a right to feel.

662
00:31:20,999 --> 00:31:22,876
You are not allowed to know the square feet and

663
00:31:22,876 --> 00:31:25,125
all the other details one.

664
00:31:25,125 --> 00:31:30,334
They have a public facing one, there's a deeper level that you have

665
00:31:30,334 --> 00:31:33,999
to sign up and be validated.

666
00:31:33,999 --> 00:31:37,125
That's what Calgary did; an admirable job of taking

667
00:31:37,125 --> 00:31:40,501
a real mess and making it good.

668
00:31:40,584 --> 00:31:42,999
Hopefully other cities will follow that lead.

669
00:31:43,542 --> 00:31:45,292
Are we done?

670
00:31:45,292 --> 00:31:46,918
MODERATOR: We're done.

671
00:31:46,918 --> 00:31:48,417
TOM KEENAN: Thanks very much.