recording-subtitles/2021/Cleaninsights/Project.kdenlive.srt

1
00:00:05,566 --> 00:00:09,734
hello everyone thank you for joining this

2
00:00:09,733 --> 00:00:13,005
session and sorry for the technical issues

3
00:00:14,400 --> 00:00:17,340
now we will welcome Nathan who is

4
00:00:17,333 --> 00:00:19,313
the founder and director of the guardian

5
00:00:19,300 --> 00:00:22,330
project an award winning open source mobile

6
00:00:22,400 --> 00:00:26,780
security collaborative in this talk using a

7
00:00:26,766 --> 00:00:30,606
clean insides neither the survey analytics based

8
00:00:30,733 --> 00:00:33,463
on matomo you will learn how to

9
00:00:33,466 --> 00:00:34,276
use the cleaning

10
00:00:34,333 --> 00:00:38,023
sides sdk from the guardian project to

11
00:00:38,100 --> 00:00:41,820
maximize maximum privacy for your users lets

12
00:00:41,800 --> 00:00:43,539
your nathan stock for more

13
00:00:44,600 --> 00:00:46,640
great thank you silver thank you lucas

14
00:00:46,633 --> 00:00:49,123
and everyone who's organized mahama camp really

15
00:00:49,133 --> 00:00:49,913
excited to

16
00:00:51,200 --> 00:00:52,880
be here with you and to start

17
00:00:53,466 --> 00:00:55,116
you have more collaboration and just to

18
00:00:55,100 --> 00:00:57,140
learn from all of you how you

19
00:00:57,133 --> 00:01:00,523
are using a moto mo to achieve

20
00:01:00,700 --> 00:01:03,370
better privacy and control of of measurement

21
00:01:03,500 --> 00:01:05,720
and analytics this is

22
00:01:07,000 --> 00:01:08,830
kind of near and dear to our

23
00:01:08,833 --> 00:01:11,173
heart at guardian project which is

24
00:01:12,300 --> 00:01:13,919
an effort that i started

25
00:01:15,066 --> 00:01:19,386
almost twelve years ago when smartphones were

26
00:01:19,400 --> 00:01:22,310
becoming very interesting and possible but i

27
00:01:22,300 --> 00:01:25,390
also quickly realized that while

28
00:01:26,766 --> 00:01:28,746
these devices that we carry around in

29
00:01:28,766 --> 00:01:31,494
our pockets are quite powerful they are

30
00:01:31,500 --> 00:01:34,892
also clearly you know tracking devices as

31
00:01:34,866 --> 00:01:38,016
well as you know the potential of

32
00:01:38,033 --> 00:01:39,533
apps which you know i've been working

33
00:01:39,533 --> 00:01:41,873
in mobile since for twenty years plus

34
00:01:41,866 --> 00:01:43,666
back to the palm pilot and apple

35
00:01:43,666 --> 00:01:46,036
newton that the fact of you know

36
00:01:46,033 --> 00:01:46,753
all of this by

37
00:01:46,766 --> 00:01:50,006
binary code running on these little devices

38
00:01:50,000 --> 00:01:51,620
and these little super computers in our

39
00:01:51,633 --> 00:01:53,103
pockets was going to cause a lot

40
00:01:53,100 --> 00:01:53,579
of trouble

41
00:01:55,533 --> 00:01:58,623
so at guardian project our focus has

42
00:01:58,600 --> 00:01:59,800
been initially

43
00:02:01,300 --> 00:02:02,770
from the beginning was saying how can

44
00:02:02,766 --> 00:02:05,346
we bring great privacy technologies that exist

45
00:02:05,366 --> 00:02:09,116
in open source and and desktop computing

46
00:02:09,199 --> 00:02:11,179
to mobile devices that we didn't even

47
00:02:11,166 --> 00:02:12,666
know if that was possible so if

48
00:02:12,666 --> 00:02:15,995
something like tor or pgp or au

49
00:02:16,000 --> 00:02:19,901
tr encryption for chat could you run

50
00:02:19,900 --> 00:02:21,070
that on a phone and people did

51
00:02:21,066 --> 00:02:22,506
i believe that you could and i

52
00:02:22,533 --> 00:02:25,083
thought he could so one of the

53
00:02:25,066 --> 00:02:27,226
things were best known for his bringing

54
00:02:27,233 --> 00:02:31,343
tour to android and more recently ios

55
00:02:32,366 --> 00:02:36,326
with orbit the version of tor that

56
00:02:36,333 --> 00:02:38,283
runs on your phone that can provide

57
00:02:38,266 --> 00:02:40,576
a vpn and through the tor anonymity

58
00:02:40,566 --> 00:02:42,183
network so there's a case where a

59
00:02:42,200 --> 00:02:46,283
very clearly connected to confidentiality and privacy

60
00:02:46,266 --> 00:02:47,706
concerns and work with one of the

61
00:02:47,733 --> 00:02:50,823
leading projects in the world and on

62
00:02:50,866 --> 00:02:52,516
on that effort another

63
00:02:52,533 --> 00:02:54,483
other piece of the puzzle that we

64
00:02:54,566 --> 00:02:56,366
quickly realized in the mobile space was

65
00:02:56,366 --> 00:03:00,776
needed was appetizer attribution beyond what you

66
00:03:00,766 --> 00:03:02,926
know the the gatekeepers of say google

67
00:03:02,933 --> 00:03:05,633
or apple or others provide and that

68
00:03:06,300 --> 00:03:10,230
much like debbie and i really is

69
00:03:10,233 --> 00:03:12,003
the debian system is really built upon

70
00:03:12,700 --> 00:03:15,369
the process of a community building and

71
00:03:15,366 --> 00:03:17,886
curating packages and distributing those and ensuring

72
00:03:17,900 --> 00:03:19,940
that you have good quality free and

73
00:03:19,933 --> 00:03:22,693
open source software we needed that for

74
00:03:22,700 --> 00:03:25,430
android and the after a project had

75
00:03:25,433 --> 00:03:27,143
already existed to be clear we did

76
00:03:27,133 --> 00:03:28,693
not found it but it was something

77
00:03:28,700 --> 00:03:30,620
that we really appreciated and wanted to

78
00:03:30,633 --> 00:03:32,853
support and have become a big part

79
00:03:32,833 --> 00:03:34,573
live through my one of my main

80
00:03:34,566 --> 00:03:38,106
collaborators collaborators hans christoph steiner

81
00:03:39,400 --> 00:03:40,570
so we're really proud of the work

82
00:03:40,566 --> 00:03:43,266
we've done and supporting that effort for

83
00:03:43,666 --> 00:03:45,256
a privacy you know an app store

84
00:03:45,266 --> 00:03:46,946
that actually tells you hey this app

85
00:03:46,933 --> 00:03:49,213
is tracking you you may not want

86
00:03:49,233 --> 00:03:52,593
to install it as an example and

87
00:03:52,566 --> 00:03:54,486
that also itself doesn't track

88
00:03:56,200 --> 00:03:58,540
the it another piece of work we've

89
00:03:58,566 --> 00:04:00,186
done in the middle there the icon

90
00:04:00,166 --> 00:04:00,496
is called

91
00:04:01,900 --> 00:04:03,550
proof mode and this is work we've

92
00:04:03,633 --> 00:04:05,223
done with witness which is a human

93
00:04:05,200 --> 00:04:09,399
rights organization focused on gathering eyewitness reports

94
00:04:09,400 --> 00:04:11,740
through video and photos and media captured

95
00:04:11,766 --> 00:04:13,266
on phones and they had the the

96
00:04:13,266 --> 00:04:15,872
vision that now there's cameras everywhere in

97
00:04:15,866 --> 00:04:17,430
every pocket and what can we do

98
00:04:17,433 --> 00:04:19,683
about that and of course many people

99
00:04:19,666 --> 00:04:20,986
there might think oh we need more

100
00:04:21,533 --> 00:04:21,804
time

101
00:04:21,800 --> 00:04:23,930
tracking and analytics and surveillance of every

102
00:04:23,933 --> 00:04:26,272
photo and every picture because we need

103
00:04:26,266 --> 00:04:28,397
to use this as evidence for justice

104
00:04:29,300 --> 00:04:32,240
but that's tricky you know we don't

105
00:04:32,233 --> 00:04:33,703
we yes maybe we want some of

106
00:04:33,700 --> 00:04:35,170
our the things we capture on our

107
00:04:35,166 --> 00:04:36,846
phones to be used to help people

108
00:04:37,466 --> 00:04:38,906
but we don't want ourselves would be

109
00:04:38,900 --> 00:04:41,150
part of the sort of panopticon surveillance

110
00:04:41,166 --> 00:04:41,886
state of

111
00:04:42,033 --> 00:04:44,643
of surveillance infrastructure in all times

112
00:04:46,500 --> 00:04:48,570
building on that we built an app

113
00:04:48,566 --> 00:04:53,246
called haven with the well known privacy

114
00:04:53,233 --> 00:04:54,553
activists edward snowden

115
00:04:55,666 --> 00:04:58,546
who again had this kind of interesting

116
00:04:58,566 --> 00:05:00,096
idea of like could you use your

117
00:05:00,066 --> 00:05:03,456
phone as a personal security device to

118
00:05:03,866 --> 00:05:06,716
monitor physical spaces like if you're a

119
00:05:06,700 --> 00:05:09,130
reporter a journalist and activist to monitor

120
00:05:09,133 --> 00:05:10,333
a place like a home or an

121
00:05:10,333 --> 00:05:12,313
office when you're not there could it

122
00:05:12,333 --> 00:05:13,733
be a kind of a watchdog or

123
00:05:13,733 --> 00:05:15,813
a you know i often think lie

124
00:05:15,800 --> 00:05:18,290
like a small robot like a r

125
00:05:18,300 --> 00:05:19,530
two d two that is on your

126
00:05:19,533 --> 00:05:21,663
side and helps monitor and protect you

127
00:05:21,666 --> 00:05:24,366
as opposed to a skynet type but

128
00:05:24,666 --> 00:05:26,856
anyway snowden ed snowden had this idea

129
00:05:26,833 --> 00:05:29,293
and we've worked through it and built

130
00:05:29,300 --> 00:05:31,040
something called haven which is quite interesting

131
00:05:31,700 --> 00:05:33,860
and lastly we've been in recent years

132
00:05:33,866 --> 00:05:35,966
working with trade unions and all

133
00:05:35,966 --> 00:05:38,846
organized labor and in many countries around

134
00:05:38,833 --> 00:05:40,663
a project called weep lock where workers

135
00:05:40,666 --> 00:05:43,666
are saying i want my data about

136
00:05:43,666 --> 00:05:45,796
my work day i don't want my

137
00:05:46,033 --> 00:05:49,273
the corporation or the you know on-demand

138
00:05:49,266 --> 00:05:50,646
platform to be the only one who's

139
00:05:50,666 --> 00:05:54,446
tracking me like there's value in me

140
00:05:54,433 --> 00:05:55,993
knowing about my own day

141
00:05:56,133 --> 00:05:58,173
and then i having the i should

142
00:05:58,166 --> 00:05:59,696
have the power to share that data

143
00:05:59,700 --> 00:06:02,670
with others to collectively organize and use

144
00:06:02,666 --> 00:06:05,636
the power of that data so these

145
00:06:05,633 --> 00:06:07,193
are all the kind of dances and

146
00:06:07,200 --> 00:06:09,690
investigations and things we've been doing around

147
00:06:09,800 --> 00:06:12,740
who should measure who should have analytics

148
00:06:12,733 --> 00:06:14,443
who should have this data and why

149
00:06:14,466 --> 00:06:16,296
and where and how when

150
00:06:16,266 --> 00:06:18,726
and you get this very either precious

151
00:06:18,766 --> 00:06:21,796
asset of gold or toxic asset of

152
00:06:22,266 --> 00:06:25,386
you know oil or how do you

153
00:06:25,533 --> 00:06:26,913
take care of it appropriately

154
00:06:28,200 --> 00:06:30,300
and yeah we're very proud of that

155
00:06:30,300 --> 00:06:32,130
work and we have a great eye

156
00:06:32,933 --> 00:06:34,373
a lot of great relationships in the

157
00:06:34,366 --> 00:06:39,166
space around privacy security open source software

158
00:06:39,600 --> 00:06:39,930
and

159
00:06:41,000 --> 00:06:42,200
this is some of some of the

160
00:06:42,200 --> 00:06:47,360
team and benjamin earhart and the top

161
00:06:47,366 --> 00:06:48,806
middle there is based in

162
00:06:49,900 --> 00:06:52,840
austria is been real lead on the

163
00:06:52,866 --> 00:06:54,396
clean insights project as well as a

164
00:06:54,366 --> 00:06:55,566
lot of our work with tour so

165
00:06:55,566 --> 00:06:56,646
i want to give him a shout

166
00:06:56,666 --> 00:06:58,766
out we also have kerry winfrey whose

167
00:06:59,266 --> 00:07:02,446
thinks about consent and the user experience

168
00:07:02,466 --> 00:07:05,076
around consent and how we engage users

169
00:07:05,066 --> 00:07:06,686
with more than just a pop-up box

170
00:07:06,666 --> 00:07:08,226
so we'll be talking about that today

171
00:07:08,966 --> 00:07:10,046
i mentioned han

172
00:07:10,066 --> 00:07:11,926
steiner who works a lot with f

173
00:07:11,933 --> 00:07:15,113
droid and debbie in and reproducible builds

174
00:07:15,666 --> 00:07:18,756
and dr gina helfrich has been a

175
00:07:18,966 --> 00:07:22,566
supporter funder mentor organizer and really thinks

176
00:07:22,566 --> 00:07:25,896
about equity and again the sustainability of

177
00:07:25,866 --> 00:07:28,236
open source and how measurement is part

178
00:07:28,266 --> 00:07:28,626
of that

179
00:07:29,733 --> 00:07:31,893
and then in learmonth who has worked

180
00:07:31,866 --> 00:07:35,256
at tour on measurement works in debian

181
00:07:35,266 --> 00:07:38,206
and works is just great on infrastructure

182
00:07:38,266 --> 00:07:41,386
devops and a lot of other really

183
00:07:41,400 --> 00:07:43,080
important pieces to this solution

184
00:07:46,533 --> 00:07:48,903
so a little while ago i got

185
00:07:48,900 --> 00:07:51,990
inspired and wrote a manifesto around clean

186
00:07:51,966 --> 00:07:55,176
insights that was inspired by the soap

187
00:07:55,233 --> 00:07:57,783
this brand of soap we have called

188
00:07:57,766 --> 00:08:00,976
dr bronner's soap which is kind of

189
00:08:00,966 --> 00:08:04,266
you know quite old maybe over nearly

190
00:08:04,700 --> 00:08:06,770
what eighty years or something and on

191
00:08:06,766 --> 00:08:10,096
the soap bottle there's always this slightly

192
00:08:10,200 --> 00:08:10,590
do

193
00:08:11,966 --> 00:08:14,846
you know creative and separate words around

194
00:08:14,833 --> 00:08:16,213
what the soap is about and i

195
00:08:16,233 --> 00:08:18,093
thought well if he can write such

196
00:08:18,066 --> 00:08:21,696
great lyrical prose about soap maybe i

197
00:08:21,700 --> 00:08:23,080
could write something around

198
00:08:24,200 --> 00:08:27,460
our work and put it on an

199
00:08:27,466 --> 00:08:29,796
interesting product that people can read every

200
00:08:29,800 --> 00:08:34,030
morning and and i did and we

201
00:08:34,033 --> 00:08:38,383
actually make coffee that is part of

202
00:08:38,400 --> 00:08:41,490
our our sort of instigation are are

203
00:08:41,466 --> 00:08:43,266
are hoped to be a catalyst and

204
00:08:44,100 --> 00:08:44,370
you can

205
00:08:44,400 --> 00:08:47,416
read this manifesto and more at the

206
00:08:47,800 --> 00:08:47,891
end

207
00:08:50,666 --> 00:08:52,316
on the link they are clean inside

208
00:08:52,300 --> 00:08:53,770
cetera it's less beans which gives you

209
00:08:53,766 --> 00:08:54,816
a sense of yeah this is a

210
00:08:54,833 --> 00:08:55,853
little bit crazy

211
00:08:57,166 --> 00:08:59,086
but it's also you know we really

212
00:08:59,100 --> 00:09:01,170
do on a change beyond just our

213
00:09:01,166 --> 00:09:03,716
software we want to change some ideas

214
00:09:03,700 --> 00:09:06,940
and thinking about measurement and analytics and

215
00:09:06,966 --> 00:09:09,696
privacy and and how it's done and

216
00:09:09,666 --> 00:09:11,286
we need a manifesto to do this

217
00:09:11,300 --> 00:09:12,800
and also we need lots of coffee

218
00:09:13,466 --> 00:09:15,686
so if you want to join us

219
00:09:15,666 --> 00:09:16,806
you'll see and you go to that

220
00:09:16,833 --> 00:09:17,253
link

221
00:09:17,566 --> 00:09:19,486
that you can actually get free coffee

222
00:09:19,600 --> 00:09:23,500
we have a distribution network nearly global

223
00:09:23,766 --> 00:09:26,826
of our insightful beans and we hope

224
00:09:26,833 --> 00:09:28,603
that every morning when you're making coffee

225
00:09:28,600 --> 00:09:30,970
he'll read our manifesto and think a

226
00:09:30,966 --> 00:09:34,416
little bit more about this and it's

227
00:09:34,433 --> 00:09:36,623
important this is really important have an

228
00:09:36,633 --> 00:09:37,405
and i'll go

229
00:09:37,433 --> 00:09:38,813
through this more in my talk about

230
00:09:38,900 --> 00:09:40,790
about why and how it impacts our

231
00:09:40,800 --> 00:09:43,050
whole lives so but free coffee if

232
00:09:43,033 --> 00:09:44,953
you sign up for a meeting as

233
00:09:44,966 --> 00:09:46,496
a follow up to this talk so

234
00:09:46,900 --> 00:09:48,280
that's the enticement

235
00:09:49,866 --> 00:09:52,566
to do this work for me and

236
00:09:52,566 --> 00:09:54,546
us as a team began a project

237
00:09:54,566 --> 00:09:56,337
i was in at harvard university am

238
00:09:56,366 --> 00:09:59,368
based in boston so it's nearby and

239
00:09:59,466 --> 00:10:00,336
i'm a fellow

240
00:10:01,400 --> 00:10:04,100
affiliate fellow now at the berkman klein

241
00:10:04,100 --> 00:10:06,230
center for internet and society there

242
00:10:07,866 --> 00:10:09,906
as sort of a twelve week workshop

243
00:10:09,966 --> 00:10:11,616
called assembly where we are meant to

244
00:10:11,633 --> 00:10:13,583
take some ideas and and put them

245
00:10:13,566 --> 00:10:16,776
into code and my colleagues on my

246
00:10:16,800 --> 00:10:19,140
team they're on my cohort were from

247
00:10:19,133 --> 00:10:25,493
apple google square and the un and

248
00:10:25,500 --> 00:10:28,050
then myself and we brainstormed

249
00:10:28,066 --> 00:10:29,806
a lot of use cases unfortunately my

250
00:10:29,800 --> 00:10:32,463
need emerged which was i have all

251
00:10:32,466 --> 00:10:34,536
of these apps and technologies and solutions

252
00:10:34,633 --> 00:10:36,433
and communities that i'm trying to help

253
00:10:36,800 --> 00:10:38,990
but i'm afraid to measure them because

254
00:10:39,233 --> 00:10:41,063
i feel like there isn't a solution

255
00:10:41,066 --> 00:10:44,126
out there yet that can and with

256
00:10:44,133 --> 00:10:46,593
uphold the privacy kind of compact we've

257
00:10:46,600 --> 00:10:48,190
made with them and that

258
00:10:48,233 --> 00:10:49,703
i i want to know more about

259
00:10:49,700 --> 00:10:51,860
what's working what's not working but i'm

260
00:10:51,900 --> 00:10:54,420
i don't want to do anything that

261
00:10:54,433 --> 00:10:57,613
would harm them and so this team

262
00:10:57,600 --> 00:10:58,980
took it seriously and it was really

263
00:10:59,000 --> 00:11:00,620
a brilliant group and kind of we

264
00:11:01,166 --> 00:11:03,506
we kickstarted so to speak the first

265
00:11:03,500 --> 00:11:07,640
implementation of cleaning sites and we also

266
00:11:07,633 --> 00:11:08,353
started by look

267
00:11:08,366 --> 00:11:09,596
looking at what was happening in the

268
00:11:09,600 --> 00:11:13,949
market and and we saw that often

269
00:11:14,333 --> 00:11:16,073
not just in apps and websites but

270
00:11:16,066 --> 00:11:18,916
you have things like the emergence of

271
00:11:18,900 --> 00:11:21,840
smart cars of you know machine learning

272
00:11:21,866 --> 00:11:24,416
based platforms where there's so much measurement

273
00:11:24,400 --> 00:11:26,920
and analytics going in order to make

274
00:11:26,933 --> 00:11:28,403
the self driving better to make the

275
00:11:28,400 --> 00:11:28,520
key

276
00:11:28,500 --> 00:11:30,420
car better to make the product better

277
00:11:30,466 --> 00:11:34,096
whatever but then so often that same

278
00:11:34,100 --> 00:11:37,340
data was being used against customers journalists

279
00:11:37,333 --> 00:11:39,881
people inquiring about the product right and

280
00:11:39,900 --> 00:11:43,050
so this was a case where a

281
00:11:43,800 --> 00:11:46,919
journalist had written an article critical of

282
00:11:46,933 --> 00:11:48,103
tesla and then

283
00:11:48,700 --> 00:11:50,710
they elon musk went and pulled out

284
00:11:50,700 --> 00:11:52,380
all of the data from that journalist

285
00:11:52,400 --> 00:11:54,480
drive and said well this is what

286
00:11:54,500 --> 00:11:55,610
you did here and this is your

287
00:11:55,600 --> 00:11:57,100
turn here and you went too fast

288
00:11:57,100 --> 00:11:59,320
here and this is how how you

289
00:11:59,333 --> 00:12:01,043
drove wrong right that was kind of

290
00:12:02,100 --> 00:12:03,360
i know there's lots of details in

291
00:12:03,366 --> 00:12:04,656
there but it's pretty shocking to me

292
00:12:04,666 --> 00:12:07,816
that the corporation could do that take

293
00:12:07,800 --> 00:12:09,329
that data and use it back against

294
00:12:09,333 --> 00:12:11,823
their critic so easily in real time

295
00:12:12,000 --> 00:12:13,110
and that's something that was meant to

296
00:12:13,133 --> 00:12:16,433
be beneficial to the customer was being

297
00:12:16,433 --> 00:12:19,583
used to harm them or to counter

298
00:12:19,566 --> 00:12:21,726
their claims and this has happened in

299
00:12:21,733 --> 00:12:22,273
a number of key

300
00:12:22,266 --> 00:12:26,196
bases around driving and accidents and other

301
00:12:26,200 --> 00:12:28,690
court cases with some of the tech

302
00:12:28,933 --> 00:12:29,593
in tesla

303
00:12:31,900 --> 00:12:33,880
there are also other cases where you

304
00:12:33,900 --> 00:12:37,800
know someone was a criminal and in

305
00:12:37,800 --> 00:12:40,439
this case there was a homeowner said

306
00:12:40,433 --> 00:12:42,443
his house burnt down but in fact

307
00:12:42,666 --> 00:12:44,586
turned out to be arson and the

308
00:12:44,600 --> 00:12:46,850
way they discovered it was arson was

309
00:12:47,233 --> 00:12:50,023
they found out he had a pacemaker

310
00:12:50,500 --> 00:12:50,830
and they

311
00:12:50,833 --> 00:12:53,473
they extracted data from his pacemaker that

312
00:12:53,466 --> 00:12:54,996
maybe he didn't even know was there

313
00:12:55,666 --> 00:12:59,506
that and to show that history wasn't

314
00:12:59,500 --> 00:13:01,090
right about was he when he was

315
00:13:01,100 --> 00:13:03,890
sleeping and this i mean this is

316
00:13:03,900 --> 00:13:07,380
like the most to stopping dystopia like

317
00:13:07,933 --> 00:13:09,853
like your you're the thing keeping you

318
00:13:09,866 --> 00:13:11,006
alive in your body

319
00:13:11,033 --> 00:13:14,153
your cybernetic implant the police are jacking

320
00:13:14,133 --> 00:13:16,323
into it and sucking out data in

321
00:13:16,333 --> 00:13:19,033
order to like you know convict you

322
00:13:19,066 --> 00:13:21,286
now he's a criminal sure but you

323
00:13:21,300 --> 00:13:24,030
know obviously there's a slippery slope and

324
00:13:24,433 --> 00:13:25,693
the more and more that we have

325
00:13:25,700 --> 00:13:29,030
technology in our bodies how and when

326
00:13:29,033 --> 00:13:30,533
can it be used against you you

327
00:13:30,533 --> 00:13:31,163
know and i know

328
00:13:31,166 --> 00:13:33,176
no as a small aspect of this

329
00:13:33,233 --> 00:13:35,843
many in the u s at least

330
00:13:36,100 --> 00:13:38,800
there's programs called wellness programs that where

331
00:13:38,800 --> 00:13:41,680
you get to a free fitbit and

332
00:13:41,700 --> 00:13:43,710
then it taps into your company's systems

333
00:13:43,700 --> 00:13:44,780
and if you wear it you get

334
00:13:44,766 --> 00:13:47,616
some coupon or credit or bonus or

335
00:13:47,633 --> 00:13:50,213
something because you're proving you're healthy right

336
00:13:50,833 --> 00:13:51,313
or

337
00:13:51,300 --> 00:13:53,909
or and yeah and so

338
00:13:55,300 --> 00:13:57,220
what next you know after that in

339
00:13:57,233 --> 00:13:59,153
the workplace and and you know we

340
00:13:59,133 --> 00:14:01,143
know already know that amazon and others

341
00:14:01,166 --> 00:14:03,476
are using kind of again things that

342
00:14:03,466 --> 00:14:05,236
are supposed to benefit you to improve

343
00:14:05,233 --> 00:14:06,553
your life to improve the service are

344
00:14:06,566 --> 00:14:07,916
being used and too often

345
00:14:09,933 --> 00:14:12,243
i know there are some highlights and

346
00:14:12,266 --> 00:14:13,946
i we had the chance you know

347
00:14:13,933 --> 00:14:16,333
through this the program at harvard to

348
00:14:16,366 --> 00:14:18,976
work with professor cynthia dork and those

349
00:14:18,966 --> 00:14:21,576
people thinking about really you know new

350
00:14:21,600 --> 00:14:23,490
kinds of maths in the order of

351
00:14:24,000 --> 00:14:25,860
do you know what end to end

352
00:14:25,833 --> 00:14:27,603
encryption and kind of the work there

353
00:14:27,600 --> 00:14:29,220
has produced and there's similar

354
00:14:29,233 --> 00:14:30,133
or work around

355
00:14:31,166 --> 00:14:34,436
statistical privacy and computing data that were

356
00:14:34,433 --> 00:14:36,203
inspired by and that we're hoping to

357
00:14:36,200 --> 00:14:38,780
work towards and we have dabbled in

358
00:14:38,766 --> 00:14:40,116
but this is sort of like you

359
00:14:40,133 --> 00:14:42,473
know there is something on the horizon

360
00:14:42,466 --> 00:14:44,716
here there's many things on the horizon

361
00:14:45,066 --> 00:14:46,536
and we were kind of able to

362
00:14:46,833 --> 00:14:49,053
tap into some of that thinking we

363
00:14:49,066 --> 00:14:50,626
are also you know solve it

364
00:14:51,300 --> 00:14:53,219
what we were doing isn't just a

365
00:14:53,500 --> 00:14:54,250
activist

366
00:14:55,633 --> 00:14:59,563
cyberpunk cypher punk you know a dream

367
00:14:59,600 --> 00:15:01,340
this is something that apple was actually

368
00:15:01,333 --> 00:15:04,063
already implementing as an example because they

369
00:15:04,066 --> 00:15:06,346
wanted to know what images were popular

370
00:15:06,333 --> 00:15:08,463
but they didn't want to log every

371
00:15:08,466 --> 00:15:10,356
single thing you were typing because they

372
00:15:10,366 --> 00:15:12,586
understood that that was a toxic asset

373
00:15:13,266 --> 00:15:14,616
and not something they wanted to hold

374
00:15:14,633 --> 00:15:15,563
onto but

375
00:15:15,566 --> 00:15:17,756
they needed to know you know just

376
00:15:17,733 --> 00:15:20,073
how popular the poop emoji was so

377
00:15:20,100 --> 00:15:21,990
they could decide where where to put

378
00:15:21,966 --> 00:15:25,176
it in the order of moses but

379
00:15:25,200 --> 00:15:27,420
that actually had a benefit without harm

380
00:15:27,600 --> 00:15:29,310
so it could work

381
00:15:31,800 --> 00:15:34,020
so you know we have we have

382
00:15:34,033 --> 00:15:36,793
we knew that the problem was monumental

383
00:15:36,833 --> 00:15:38,783
we knew that like there is some

384
00:15:38,766 --> 00:15:40,956
possibilities and i should i should also

385
00:15:40,966 --> 00:15:43,456
say you know back then with you

386
00:15:43,466 --> 00:15:45,416
know p wc and and matomo like

387
00:15:45,400 --> 00:15:47,620
we were always inspired by this work

388
00:15:47,633 --> 00:15:50,393
in this community building open source self

389
00:15:50,400 --> 00:15:50,610
hosts

390
00:15:50,633 --> 00:15:53,705
it analytics that you know kind of

391
00:15:53,700 --> 00:15:55,230
fit in the world of the web

392
00:15:55,233 --> 00:15:57,723
that we believe in and in terms

393
00:15:57,733 --> 00:16:00,343
of decentralized technologies and self hosting and

394
00:16:00,333 --> 00:16:01,923
we always like look these these guys

395
00:16:01,933 --> 00:16:03,253
are doing it this this team is

396
00:16:03,800 --> 00:16:05,660
producing products that people are starting to

397
00:16:05,666 --> 00:16:07,346
adopt you know and and could we

398
00:16:07,333 --> 00:16:09,283
build something like that or build on

399
00:16:09,300 --> 00:16:10,680
it right into the seeds or already

400
00:16:10,666 --> 00:16:10,756
have

401
00:16:10,766 --> 00:16:13,196
they're for connecting the work we're doing

402
00:16:13,200 --> 00:16:14,880
with with the matomo community

403
00:16:16,066 --> 00:16:16,546
now

404
00:16:17,900 --> 00:16:19,550
most recently you know in the last

405
00:16:19,533 --> 00:16:22,023
year we've also seen that there's continued

406
00:16:22,033 --> 00:16:24,283
to be kind of big gaffes and

407
00:16:24,866 --> 00:16:28,256
missteps and problems with even beloved software

408
00:16:28,233 --> 00:16:31,053
and open source community i love audacity

409
00:16:31,066 --> 00:16:32,266
i'm a huge fan of use it

410
00:16:32,266 --> 00:16:34,786
for a long time and you know

411
00:16:34,800 --> 00:16:35,490
we saw

412
00:16:37,033 --> 00:16:39,463
a real misstep by this new parent

413
00:16:39,466 --> 00:16:43,006
company that acquired the code or the

414
00:16:43,000 --> 00:16:45,100
team and and all the details but

415
00:16:45,100 --> 00:16:47,680
the point is that there's still a

416
00:16:47,700 --> 00:16:49,470
lot of lack of awareness around like

417
00:16:49,466 --> 00:16:50,846
why would it be bad to just

418
00:16:50,833 --> 00:16:53,233
drop in you know perma cookies in

419
00:16:53,233 --> 00:16:55,483
your new ids and commercial toolkits into

420
00:16:55,500 --> 00:16:56,850
this product like that's what we do

421
00:16:56,833 --> 00:16:59,323
dude that's what everybody does right there

422
00:16:59,333 --> 00:17:00,216
still a basic

423
00:17:01,433 --> 00:17:05,633
huge actually chaz and across around what

424
00:17:05,666 --> 00:17:07,406
what is what is ok what is

425
00:17:07,400 --> 00:17:09,290
okay to do you know why shouldn't

426
00:17:09,300 --> 00:17:10,530
we and even in the human rights

427
00:17:10,500 --> 00:17:13,139
community and even in you know the

428
00:17:13,166 --> 00:17:14,995
harbor community but people just drop in

429
00:17:15,000 --> 00:17:16,441
google analytics and i know that's a

430
00:17:16,433 --> 00:17:18,113
thing that you all are working on

431
00:17:18,099 --> 00:17:19,478
and probably have a lot more experience

432
00:17:19,500 --> 00:17:21,510
and how to get people to understand

433
00:17:21,500 --> 00:17:21,561
the

434
00:17:21,566 --> 00:17:23,426
the other possibilities and and what can

435
00:17:23,400 --> 00:17:25,380
be done there and that's part of

436
00:17:25,400 --> 00:17:27,709
our work especially in open source community

437
00:17:29,066 --> 00:17:32,725
and we also see this continued kind

438
00:17:32,700 --> 00:17:36,751
of an attack on users around conflating

439
00:17:36,800 --> 00:17:41,330
usability improvements with advertiser tracking for advertising

440
00:17:41,333 --> 00:17:42,353
and i think that is a big

441
00:17:42,366 --> 00:17:44,196
thing that gets still combined quite a

442
00:17:44,200 --> 00:17:46,060
bit where you said oh we need

443
00:17:46,066 --> 00:17:47,746
your help to do x y and

444
00:17:47,733 --> 00:17:48,903
z and make this better

445
00:17:49,233 --> 00:17:50,613
but then that data is you know

446
00:17:50,633 --> 00:17:54,383
combined for other purposes and again if

447
00:17:54,366 --> 00:17:56,706
we are building software for human rights

448
00:17:56,700 --> 00:17:59,850
humanitarian purposes and then this kind of

449
00:18:00,466 --> 00:18:02,536
gets introduced in the process where we

450
00:18:02,533 --> 00:18:04,243
say oh you have to use this

451
00:18:04,700 --> 00:18:06,080
other tool and then it's going to

452
00:18:06,066 --> 00:18:09,066
track you it's it overwhelming kind of

453
00:18:09,133 --> 00:18:13,093
number of datapoints start being collected on

454
00:18:13,100 --> 00:18:15,320
people we're trying to help and you

455
00:18:15,300 --> 00:18:17,490
know i don't i kudos to apple

456
00:18:17,500 --> 00:18:19,270
for what they've been you know some

457
00:18:19,266 --> 00:18:20,316
of the moves they've made have been

458
00:18:20,333 --> 00:18:23,903
really interesting with injecting into the user

459
00:18:23,900 --> 00:18:26,780
experience around consent a new model you

460
00:18:26,766 --> 00:18:28,626
know i think people are starting to

461
00:18:28,633 --> 00:18:29,233
respond

462
00:18:29,700 --> 00:18:31,230
and it's interesting to see how it's

463
00:18:31,233 --> 00:18:33,753
impacted companies like facebook

464
00:18:37,133 --> 00:18:39,203
and and then you know with with

465
00:18:39,200 --> 00:18:41,990
the pandemic you know we saw a

466
00:18:42,000 --> 00:18:44,536
lot of emerge and in recent years

467
00:18:44,833 --> 00:18:47,083
emergence of smart kind of civic and

468
00:18:47,100 --> 00:18:50,129
health technology you know be it and

469
00:18:50,700 --> 00:18:54,840
measuring people in cities measuring people around

470
00:18:54,866 --> 00:18:55,166
each

471
00:18:55,166 --> 00:18:58,496
other you know having more data about

472
00:18:58,533 --> 00:19:02,223
you know covert and testing and having

473
00:19:03,066 --> 00:19:05,436
more public infrastructure like in new york

474
00:19:05,433 --> 00:19:07,563
city these wifi hotspots that are meant

475
00:19:07,566 --> 00:19:10,386
to provide people without access to the

476
00:19:10,400 --> 00:19:13,490
internet free wifi and chromebooks and all

477
00:19:13,500 --> 00:19:15,330
of this stuff and you know

478
00:19:15,333 --> 00:19:17,103
we we've been through a lot and

479
00:19:17,133 --> 00:19:18,213
you know i hope all of you

480
00:19:18,733 --> 00:19:20,001
you know are are doing well and

481
00:19:20,000 --> 00:19:21,410
your families are are well and that

482
00:19:21,533 --> 00:19:22,673
you know we're all on the road

483
00:19:22,666 --> 00:19:24,946
to a better brighter day but there's

484
00:19:24,933 --> 00:19:27,243
been a lot of kind of changes

485
00:19:27,300 --> 00:19:31,080
and again what what is happening in

486
00:19:31,066 --> 00:19:34,336
our society around infrastructure and you know

487
00:19:34,333 --> 00:19:35,473
on the one hand there's there's

488
00:19:35,466 --> 00:19:38,016
as horrific security which different topic for

489
00:19:38,033 --> 00:19:39,713
most of these things but on the

490
00:19:39,700 --> 00:19:42,760
other hand you know again what is

491
00:19:42,766 --> 00:19:44,656
what are you collecting and why and

492
00:19:44,666 --> 00:19:46,646
when what data is collected how long

493
00:19:46,633 --> 00:19:47,682
do you retain it and all of

494
00:19:47,700 --> 00:19:49,741
this that i think is is very

495
00:19:49,733 --> 00:19:51,413
different country to country as well so

496
00:19:51,400 --> 00:19:52,720
i want to recognize that based on

497
00:19:52,733 --> 00:19:55,643
different regulations but there's

498
00:19:56,033 --> 00:19:57,653
still that vision like if we just

499
00:19:57,633 --> 00:20:00,123
knew everything everything would be okay so

500
00:20:02,200 --> 00:20:04,570
so you know back to arc our

501
00:20:04,566 --> 00:20:06,816
core story and we have a lot

502
00:20:06,833 --> 00:20:09,233
of these users that we want to

503
00:20:09,233 --> 00:20:11,663
help authentically to make our software that

504
00:20:11,666 --> 00:20:13,136
doesn't track them it doesn't harm the

505
00:20:13,133 --> 00:20:15,443
better but we don't and we haven't

506
00:20:15,433 --> 00:20:16,723
been in and we needed to figure

507
00:20:16,733 --> 00:20:18,353
out a way to do that and

508
00:20:18,333 --> 00:20:20,763
so there's a lot of others that

509
00:20:20,766 --> 00:20:21,306
are in the same

510
00:20:22,200 --> 00:20:25,200
quandary and that we work with and

511
00:20:25,200 --> 00:20:26,460
that we've been trying to provide a

512
00:20:26,466 --> 00:20:28,296
solution for in our space and i'm

513
00:20:28,300 --> 00:20:29,560
here today to kind of share that

514
00:20:29,566 --> 00:20:31,846
work and see maybe there's broader adoption

515
00:20:31,833 --> 00:20:32,553
possibility

516
00:20:33,900 --> 00:20:36,600
so as i said the tor project

517
00:20:37,366 --> 00:20:39,046
has actually done a lot of thinking

518
00:20:39,066 --> 00:20:41,976
about this and they are a big

519
00:20:41,966 --> 00:20:44,816
inspiration with their metrics portal which is

520
00:20:44,833 --> 00:20:47,713
completely public and it's so interesting that

521
00:20:47,700 --> 00:20:50,250
an anonymity system actually has the best

522
00:20:50,433 --> 00:20:52,533
measurement and analytics system as part of

523
00:20:52,533 --> 00:20:53,913
it but they fee

524
00:20:53,900 --> 00:20:55,040
figured out how to do it they

525
00:20:55,066 --> 00:20:57,256
found that compromise and that way to

526
00:20:57,500 --> 00:21:01,040
do it without cock their users and

527
00:21:01,066 --> 00:21:02,746
they've communicate and it's a kind of

528
00:21:02,733 --> 00:21:05,313
a key very transparent piece of their

529
00:21:05,300 --> 00:21:08,360
system to say his tour working how

530
00:21:08,366 --> 00:21:10,136
many people are using it what's the

531
00:21:10,133 --> 00:21:12,533
throughput in a where is our infrastructure

532
00:21:12,533 --> 00:21:14,063
located so all of these things

533
00:21:14,100 --> 00:21:20,250
r and r there and i know

534
00:21:20,266 --> 00:21:21,736
they continue to do this and really

535
00:21:21,733 --> 00:21:23,623
have been excellent there's others like the

536
00:21:23,600 --> 00:21:27,800
open observatory of network interference and other

537
00:21:27,800 --> 00:21:31,010
groups doing measurement that we take inspiration

538
00:21:31,033 --> 00:21:33,913
from we also did an outreach in

539
00:21:33,900 --> 00:21:34,230
a row

540
00:21:34,233 --> 00:21:36,483
a research project with that is published

541
00:21:36,466 --> 00:21:38,146
on clean insights dot org and through

542
00:21:38,166 --> 00:21:42,396
partner sites with groups like keepass cubes

543
00:21:42,700 --> 00:21:46,960
tunnelbear least authority siphon umbrella a bunch

544
00:21:46,966 --> 00:21:48,766
of different projects to say hey how

545
00:21:48,766 --> 00:21:50,176
do you feel what are you doing

546
00:21:50,166 --> 00:21:54,366
with measurement and how are you do

547
00:21:56,500 --> 00:21:58,600
so if we could offer more what

548
00:21:58,666 --> 00:22:00,166
you know would you take it and

549
00:22:00,466 --> 00:22:02,086
this is the big question really and

550
00:22:02,066 --> 00:22:03,746
how they interact with their users and

551
00:22:03,766 --> 00:22:06,196
what they expect and so we were

552
00:22:06,166 --> 00:22:07,846
excited to collaborate and there's a lot

553
00:22:07,866 --> 00:22:09,786
to learn from that report and and

554
00:22:09,766 --> 00:22:12,436
part of it was also that fed

555
00:22:12,466 --> 00:22:14,566
into a symposium we held last year

556
00:22:14,566 --> 00:22:14,686
your

557
00:22:16,366 --> 00:22:18,706
and we have told another one next

558
00:22:18,733 --> 00:22:21,073
year that really brought us together into

559
00:22:21,066 --> 00:22:22,866
a lot of the the questions around

560
00:22:23,533 --> 00:22:27,343
consent and kind of what should they

561
00:22:27,333 --> 00:22:29,133
know and what should you know and

562
00:22:29,133 --> 00:22:32,013
how should you measure it so out

563
00:22:32,000 --> 00:22:33,770
of all of this finally getting to

564
00:22:33,766 --> 00:22:35,866
the point here do we

565
00:22:35,900 --> 00:22:38,630
decided to build some software and i

566
00:22:39,133 --> 00:22:41,863
you know i mentioned my collaborators earlier

567
00:22:42,166 --> 00:22:44,176
and we said look can we actually

568
00:22:44,166 --> 00:22:46,206
build something that you know and when

569
00:22:46,200 --> 00:22:48,360
we had tinkered since twenty seventeen on

570
00:22:48,366 --> 00:22:50,616
different little pieces and trying things but

571
00:22:50,900 --> 00:22:52,400
i think we struck get a much

572
00:22:52,400 --> 00:22:54,830
clearer idea in a specification of what

573
00:22:55,366 --> 00:22:56,086
we would use

574
00:22:56,066 --> 00:22:59,096
use in in a toolkit for measurements

575
00:22:59,166 --> 00:23:01,956
and how what we would need as

576
00:23:01,966 --> 00:23:04,126
part of that and a big part

577
00:23:04,100 --> 00:23:05,360
of it was just this idea again

578
00:23:05,366 --> 00:23:07,316
i mentioned the idea of toxic assets

579
00:23:07,333 --> 00:23:09,613
and liability and how do we separate

580
00:23:09,633 --> 00:23:12,213
that risk from the actual thing we

581
00:23:12,200 --> 00:23:14,510
want to know and so our model

582
00:23:14,533 --> 00:23:16,243
is not to log as

583
00:23:16,233 --> 00:23:17,943
everything and then figure it out later

584
00:23:17,933 --> 00:23:19,913
but it's to ask specific questions and

585
00:23:19,933 --> 00:23:24,253
measure specific things and do edge client

586
00:23:24,233 --> 00:23:27,353
side processing in browsers in mobile apps

587
00:23:27,500 --> 00:23:29,570
and then submit the data in a

588
00:23:29,566 --> 00:23:32,806
way that is as clean as possible

589
00:23:33,733 --> 00:23:34,243
so

590
00:23:35,466 --> 00:23:37,446
do you know some of the basic

591
00:23:37,433 --> 00:23:40,324
kind of guiding principles is around that

592
00:23:40,333 --> 00:23:43,322
but we have are displayed here and

593
00:23:43,766 --> 00:23:46,946
and we work through a variety of

594
00:23:46,966 --> 00:23:49,636
mechanisms to achieve these things about aggregating

595
00:23:49,633 --> 00:23:53,443
and averaging and pointed focus data collection

596
00:23:53,866 --> 00:23:55,606
and measurement wind

597
00:23:55,600 --> 00:23:57,820
does wear of time so you don't

598
00:23:57,800 --> 00:23:59,540
get specific timestamps

599
00:24:00,666 --> 00:24:02,766
getting consent to opt in and how

600
00:24:02,766 --> 00:24:04,476
we fish explain what's going to be

601
00:24:04,466 --> 00:24:06,086
measured this is all part of what

602
00:24:06,166 --> 00:24:07,606
insights is designed to do

603
00:24:08,933 --> 00:24:09,412
i

604
00:24:10,666 --> 00:24:12,136
and a bit of a different view

605
00:24:12,166 --> 00:24:14,206
of this i should mention at this

606
00:24:14,200 --> 00:24:15,879
point we decided very early we don't

607
00:24:15,866 --> 00:24:17,066
want to build our own back end

608
00:24:17,066 --> 00:24:18,536
system we want to focus on our

609
00:24:18,533 --> 00:24:21,893
client side st ks and so we

610
00:24:21,900 --> 00:24:24,780
chose moto mo as the sort of

611
00:24:24,766 --> 00:24:27,226
infrastructure we wanted to do

612
00:24:28,500 --> 00:24:31,230
integrate with the analytics and visualization and

613
00:24:31,233 --> 00:24:33,333
kind of long term or short term

614
00:24:33,333 --> 00:24:37,113
storage of data and it's been great

615
00:24:37,100 --> 00:24:40,520
and it's also interesting because many organizations

616
00:24:40,566 --> 00:24:42,636
want to use moto mo for their

617
00:24:42,700 --> 00:24:44,740
their basic web analytics for their webpage

618
00:24:44,733 --> 00:24:46,443
for their marketing still and that's fine

619
00:24:46,800 --> 00:24:48,630
what we're talking about our application

620
00:24:48,633 --> 00:24:51,663
ns and infrastructure specific interactions and so

621
00:24:51,666 --> 00:24:53,856
it actually works well together to say

622
00:24:53,866 --> 00:24:55,756
lt i set up a tomo will

623
00:24:55,733 --> 00:24:58,073
run it for you dropping the you

624
00:24:58,066 --> 00:24:59,896
know the javascript in your website let's

625
00:24:59,933 --> 00:25:01,583
tune it for privacy there's a lot

626
00:25:01,600 --> 00:25:03,280
of great settings hematoma we love for

627
00:25:03,266 --> 00:25:05,246
that and then for your app for

628
00:25:05,266 --> 00:25:07,426
your other things we'll use clean insights

629
00:25:07,433 --> 00:25:08,783
on top of that and they

630
00:25:08,833 --> 00:25:10,813
get this really great combined view of

631
00:25:11,033 --> 00:25:12,330
kind of what they're used to with

632
00:25:12,333 --> 00:25:14,133
something new and we've been really you

633
00:25:14,133 --> 00:25:16,863
know happy about that and so do

634
00:25:17,700 --> 00:25:18,870
a lot of that back and i'll

635
00:25:18,866 --> 00:25:20,336
talk about the privacy proxy in a

636
00:25:20,333 --> 00:25:23,963
moment but we have this client side

637
00:25:23,966 --> 00:25:26,996
st ks that are configured and have

638
00:25:27,066 --> 00:25:28,956
measurements of events and views

639
00:25:29,000 --> 00:25:30,440
that have a bit of processing in

640
00:25:30,466 --> 00:25:32,416
the storage queue and then are sent

641
00:25:32,400 --> 00:25:35,580
at a certain period back to this

642
00:25:35,566 --> 00:25:37,396
privacy proxy we built and then that

643
00:25:37,466 --> 00:25:38,516
goes into matomo

644
00:25:42,233 --> 00:25:44,243
so as of today on git lab

645
00:25:44,266 --> 00:25:46,366
and unclean insights dot org we have

646
00:25:46,800 --> 00:25:50,460
actual st ks that have shipped and

647
00:25:51,100 --> 00:25:55,030
are very exciting to to proud that

648
00:25:55,033 --> 00:25:57,193
they're there and we're integrating in applications

649
00:25:57,166 --> 00:25:59,896
kenneth as we speak for

650
00:25:59,933 --> 00:26:01,883
lots of languages and i think we

651
00:26:01,900 --> 00:26:04,960
actually also have the rust sdk is

652
00:26:04,966 --> 00:26:06,136
our newest one you know and i

653
00:26:06,133 --> 00:26:09,463
know rust is emerging in many environments

654
00:26:09,466 --> 00:26:11,056
is like the most secure way you

655
00:26:11,033 --> 00:26:13,883
can build something that is you know

656
00:26:14,066 --> 00:26:16,706
a binary kind of application but we

657
00:26:16,700 --> 00:26:19,670
support server applications through python we support

658
00:26:19,700 --> 00:26:20,060
apple

659
00:26:20,066 --> 00:26:22,706
full android javascript so there's lots of

660
00:26:22,700 --> 00:26:26,450
ways to start dabbling in it and

661
00:26:26,466 --> 00:26:29,136
using our code in different environments and

662
00:26:29,133 --> 00:26:30,504
then i'll talk a bit more about

663
00:26:30,566 --> 00:26:32,246
potomac proxy so this is all on

664
00:26:32,266 --> 00:26:35,146
our website everything's to get lab published

665
00:26:35,133 --> 00:26:37,683
in the places you'd expect for dependencies

666
00:26:38,100 --> 00:26:39,990
and so on so she very easy

667
00:26:40,000 --> 00:26:40,240
to end

668
00:26:40,333 --> 00:26:40,781
great

669
00:26:42,000 --> 00:26:44,100
and a piece early on that we

670
00:26:44,100 --> 00:26:46,170
realized we needed to achieve some of

671
00:26:46,166 --> 00:26:48,596
our goals was a privacy proxy in

672
00:26:48,600 --> 00:26:51,780
front of mahama and this is what

673
00:26:51,766 --> 00:26:55,366
our sdk speaks to in order to

674
00:26:55,633 --> 00:26:55,963
do

675
00:26:57,966 --> 00:26:59,616
interact with the data in the way

676
00:26:59,600 --> 00:27:01,424
we collected and then put that into

677
00:27:01,500 --> 00:27:03,990
matomo in the appropriate way so this

678
00:27:04,000 --> 00:27:06,850
is this isn't a great slide but

679
00:27:07,066 --> 00:27:09,256
this is really the information from the

680
00:27:09,333 --> 00:27:10,683
the the git lab in the read

681
00:27:10,666 --> 00:27:13,516
me and a benjamin earhart who's here

682
00:27:14,300 --> 00:27:16,490
design most of this and but we

683
00:27:16,500 --> 00:27:17,190
had certain requests

684
00:27:17,200 --> 00:27:19,240
garments and the idea was we would

685
00:27:20,033 --> 00:27:21,713
you know we we kind of aggregate

686
00:27:21,766 --> 00:27:23,566
records together so they're not always being

687
00:27:23,566 --> 00:27:25,876
logged in real time as an example

688
00:27:26,733 --> 00:27:28,533
we also wanted to filter some of

689
00:27:28,533 --> 00:27:30,963
the headers from mobile devices and things

690
00:27:30,966 --> 00:27:32,616
and so that we could really control

691
00:27:32,600 --> 00:27:33,860
some of those things before they ever

692
00:27:33,866 --> 00:27:37,346
went into matomo and and recognizing

693
00:27:37,366 --> 00:27:39,406
that in some case the number one

694
00:27:39,466 --> 00:27:42,076
with all the materials went maybe as

695
00:27:42,100 --> 00:27:45,700
precisely as they wanted and yeah and

696
00:27:45,700 --> 00:27:48,310
it's a pretty simple layer but it

697
00:27:48,300 --> 00:27:50,160
was great that we had them tomorrow

698
00:27:50,166 --> 00:27:51,726
a poi to work with them that

699
00:27:51,733 --> 00:27:52,993
we could build this layer that we

700
00:27:52,966 --> 00:27:56,446
needed so again it's about flexibility control

701
00:27:56,800 --> 00:27:57,520
and it was awesome

702
00:27:57,566 --> 00:28:00,116
that that this was possible and

703
00:28:02,000 --> 00:28:03,140
been working out very well

704
00:28:06,466 --> 00:28:09,106
so when you work with and clean

705
00:28:09,133 --> 00:28:12,463
insights the steps are that first you

706
00:28:13,266 --> 00:28:15,126
set up a campaign and this is

707
00:28:15,133 --> 00:28:18,913
through coups de son or configuration file

708
00:28:18,933 --> 00:28:21,363
that defines kind of the measurement period

709
00:28:21,933 --> 00:28:23,613
and some

710
00:28:25,866 --> 00:28:28,236
local persistence how often to measure when

711
00:28:28,233 --> 00:28:30,783
to start and when to end and

712
00:28:31,000 --> 00:28:32,530
this can be set very broad or

713
00:28:32,533 --> 00:28:34,423
can be set very narrow and so

714
00:28:34,400 --> 00:28:35,570
one of the ideas as you might

715
00:28:35,566 --> 00:28:37,156
say to users i'm just going to

716
00:28:37,166 --> 00:28:39,716
measure you for two weeks we're doing

717
00:28:39,733 --> 00:28:41,863
a study for two weeks do you

718
00:28:41,866 --> 00:28:43,186
want to join it's going to be

719
00:28:43,166 --> 00:28:44,546
focused on improve

720
00:28:44,566 --> 00:28:46,756
seeing this feature and you can join

721
00:28:46,733 --> 00:28:49,583
and participate in this study that's a

722
00:28:49,600 --> 00:28:51,879
big key concept here which is not

723
00:28:51,866 --> 00:28:54,566
sort of endless measurement and we think

724
00:28:54,566 --> 00:28:57,296
people are more likely to opt in

725
00:28:57,300 --> 00:28:59,130
and say sure i'm excited about that

726
00:28:59,133 --> 00:29:01,293
if they have that ability so then

727
00:29:01,300 --> 00:29:02,860
we get consent and we have some

728
00:29:02,866 --> 00:29:04,006
various you x i'm going to get

729
00:29:04,000 --> 00:29:04,360
to

730
00:29:04,733 --> 00:29:06,053
quickly cause i'm running out of time

731
00:29:06,733 --> 00:29:08,863
and then you choose to measure specific

732
00:29:08,866 --> 00:29:11,926
events with custom data or specific views

733
00:29:12,333 --> 00:29:13,653
some of this relates to what you

734
00:29:13,666 --> 00:29:14,656
might be used to in them and

735
00:29:14,633 --> 00:29:17,603
tomo st ks around events abused because

736
00:29:17,700 --> 00:29:19,050
we knew the data was ending up

737
00:29:19,066 --> 00:29:20,236
there in the end so we wanted

738
00:29:20,233 --> 00:29:21,553
to be able to map it into

739
00:29:22,066 --> 00:29:23,686
some of what you would see in

740
00:29:23,833 --> 00:29:24,883
the image

741
00:29:24,866 --> 00:29:26,606
item on the backend right so where

742
00:29:27,133 --> 00:29:28,993
it's you know there there is that

743
00:29:28,966 --> 00:29:31,366
heritage there and thinking about the data

744
00:29:31,400 --> 00:29:33,380
but how we process it and it

745
00:29:33,400 --> 00:29:35,770
gets there over time so you might

746
00:29:35,766 --> 00:29:38,076
do a bunch of measurements of someone

747
00:29:38,066 --> 00:29:41,036
clicking on a button or interacting with

748
00:29:41,033 --> 00:29:42,563
something and then once a week that

749
00:29:42,566 --> 00:29:45,026
data is aggregated as account couch

750
00:29:45,033 --> 00:29:46,413
sent and sent to the server as

751
00:29:46,400 --> 00:29:48,950
opposed to off you know going constantly

752
00:29:48,966 --> 00:29:49,686
to the back end

753
00:29:51,233 --> 00:29:52,943
so clean insights dot org has a

754
00:29:52,966 --> 00:29:54,796
lot more information on this and we

755
00:29:54,766 --> 00:29:57,196
are also doing a workshop tomorrow morning

756
00:29:57,233 --> 00:29:59,723
that do you know if you really

757
00:29:59,733 --> 00:30:00,723
want to implement we're going to be

758
00:30:00,733 --> 00:30:02,023
ready to talk to you about your

759
00:30:02,000 --> 00:30:04,400
specific needs so and and i'm always

760
00:30:04,400 --> 00:30:04,850
available

761
00:30:06,000 --> 00:30:07,860
a big part of this in our

762
00:30:07,866 --> 00:30:09,816
apps was also this idea of consent

763
00:30:09,900 --> 00:30:11,910
and we've spent a lot in consent

764
00:30:11,900 --> 00:30:13,970
user experience design a lot of energy

765
00:30:14,466 --> 00:30:17,316
around the messaging contribute for the number

766
00:30:17,300 --> 00:30:19,310
of time you know we'll be doing

767
00:30:19,333 --> 00:30:21,283
these things to gather these pieces of

768
00:30:21,266 --> 00:30:23,126
data and here's how you'll benefit and

769
00:30:23,133 --> 00:30:24,843
we'll reach back to you and so

770
00:30:24,833 --> 00:30:26,063
really make it a pi

771
00:30:26,600 --> 00:30:29,960
tory co-design process to improve your software

772
00:30:29,966 --> 00:30:32,276
or your site in a way that's

773
00:30:32,266 --> 00:30:34,936
transparent and collaborative this is key for

774
00:30:34,933 --> 00:30:36,823
us and we think it's an important

775
00:30:37,133 --> 00:30:39,923
model moving forward you know you again

776
00:30:39,900 --> 00:30:40,950
you could say we're going to measure

777
00:30:40,966 --> 00:30:42,496
you for a year or for a

778
00:30:42,500 --> 00:30:45,230
month or longer time periods but i

779
00:30:45,500 --> 00:30:46,671
do want to kind of have it

780
00:30:46,666 --> 00:30:48,076
an experience where they you know it's

781
00:30:48,066 --> 00:30:48,216
like

782
00:30:50,000 --> 00:30:51,470
people in the u s there's a

783
00:30:51,533 --> 00:30:53,213
thing called nielsen ratings where it used

784
00:30:53,200 --> 00:30:54,820
to be super excited in the old

785
00:30:54,833 --> 00:30:56,453
days of television to be a nielsen

786
00:30:56,433 --> 00:30:58,893
family because they wanted to know what

787
00:30:58,900 --> 00:31:00,880
you cared about and which tv shows

788
00:31:00,866 --> 00:31:02,126
were good and you had a box

789
00:31:02,133 --> 00:31:03,483
in your house and it felt like

790
00:31:04,200 --> 00:31:06,330
everyone wanted to be that family and

791
00:31:06,333 --> 00:31:07,743
i think we're trying to recapture some

792
00:31:07,766 --> 00:31:09,296
of that positivity around

793
00:31:09,833 --> 00:31:12,293
and providing feedback to services

794
00:31:14,166 --> 00:31:17,915
and yeah ultimately there's an idea of

795
00:31:17,933 --> 00:31:21,354
why what you're gathering and why and

796
00:31:21,333 --> 00:31:22,773
that we want to tap into so

797
00:31:22,866 --> 00:31:24,306
i'm going to keep going so we

798
00:31:24,300 --> 00:31:26,700
have a bunch of designs figma files

799
00:31:27,366 --> 00:31:29,814
blog posts through our partner okay thanks

800
00:31:30,133 --> 00:31:32,383
carry on our team is that's her

801
00:31:32,366 --> 00:31:33,446
design studio and

802
00:31:33,533 --> 00:31:36,053
really has done outstanding work here and

803
00:31:36,033 --> 00:31:37,563
we're actually building all of this into

804
00:31:37,600 --> 00:31:40,180
multiple applications like this i mentioned earlier

805
00:31:40,833 --> 00:31:45,363
and f droid themselves have already implemented

806
00:31:46,766 --> 00:31:48,326
a piece of this though we haven't

807
00:31:48,333 --> 00:31:50,553
fully done the consent you user experience

808
00:31:51,466 --> 00:31:52,452
but if you go to the after

809
00:31:52,466 --> 00:31:53,996
a blog there's a post on clean

810
00:31:53,966 --> 00:31:56,396
insights and that the ideas was we

811
00:31:56,400 --> 00:31:59,130
wanted to understand the insight was how

812
00:31:59,133 --> 00:32:00,963
often do people fail to install an

813
00:32:00,966 --> 00:32:02,766
app they choose and then one of

814
00:32:02,766 --> 00:32:05,166
the most popular apps this week across

815
00:32:05,166 --> 00:32:06,276
all repositories

816
00:32:06,633 --> 00:32:08,043
and that felt like something that they

817
00:32:08,066 --> 00:32:10,616
were comfortable using clean insights to measure

818
00:32:10,900 --> 00:32:12,700
as people opted in and that would

819
00:32:12,700 --> 00:32:14,410
be that people would be excited maybe

820
00:32:14,400 --> 00:32:16,350
to share that right and so those

821
00:32:16,366 --> 00:32:17,836
are the idea of insights that we

822
00:32:17,833 --> 00:32:19,723
feel are clean that we are able

823
00:32:19,733 --> 00:32:20,633
to generate there

824
00:32:22,366 --> 00:32:24,856
so you know in the future there's

825
00:32:24,866 --> 00:32:26,186
a lot more we want to do

826
00:32:26,166 --> 00:32:29,106
with more advanced kind of solutions and

827
00:32:29,100 --> 00:32:31,800
more filters and how to to batch

828
00:32:31,800 --> 00:32:35,400
data we're also looking at secure multiparty

829
00:32:35,400 --> 00:32:37,890
computation and some of the interesting work

830
00:32:37,900 --> 00:32:40,540
there with kind of you know how

831
00:32:40,533 --> 00:32:41,853
to headaches expert

832
00:32:41,866 --> 00:32:45,076
and what we're doing into larger datasets

833
00:32:45,566 --> 00:32:48,026
and have solutions for a broader market

834
00:32:49,200 --> 00:32:51,960
and again free coffee

835
00:32:54,200 --> 00:32:56,720
so good so delicious so if you

836
00:32:56,700 --> 00:32:59,670
come to our workshop tomorrow i think

837
00:32:59,666 --> 00:33:02,096
you got free coffee sent to you

838
00:33:02,366 --> 00:33:05,366
and then so good and

839
00:33:07,033 --> 00:33:08,443
you know but this is also if

840
00:33:08,433 --> 00:33:09,663
you please if you go there you'll

841
00:33:09,666 --> 00:33:10,776
see you know it's a chance for

842
00:33:10,766 --> 00:33:12,386
us to connect and we really wanted

843
00:33:12,433 --> 00:33:14,023
to we have so much to learn

844
00:33:14,033 --> 00:33:17,633
from you and from people that i

845
00:33:18,033 --> 00:33:19,983
have already been doing measurement are already

846
00:33:20,000 --> 00:33:21,830
implementing things like madame i've been through

847
00:33:21,833 --> 00:33:23,633
this or who have other thoughts or

848
00:33:23,633 --> 00:33:25,613
other ways that they've solve the issues

849
00:33:25,600 --> 00:33:26,380
we have brought up

850
00:33:26,866 --> 00:33:28,876
so we're really wanting to get that

851
00:33:28,933 --> 00:33:31,592
that feedback for sure and and

852
00:33:32,866 --> 00:33:34,066
hope my slight is a little messed

853
00:33:34,066 --> 00:33:35,926
up but i'm you know and eight

854
00:33:35,933 --> 00:33:37,403
fr eight nate freight

855
00:33:38,500 --> 00:33:41,680
in many places on matrix nathan a

856
00:33:41,666 --> 00:33:43,946
garden project at info and clean insights

857
00:33:43,966 --> 00:33:48,068
dot org and benjamin earhart is tl

858
00:33:48,066 --> 00:33:50,856
a he's often around in various rooms

859
00:33:51,366 --> 00:33:53,466
and yeah so i think that's my

860
00:33:53,466 --> 00:33:58,656
talk for today and and we have

861
00:33:58,666 --> 00:34:00,796
have time for questions i think because

862
00:34:00,800 --> 00:34:02,420
we did start a little bit late

863
00:34:03,366 --> 00:34:04,926
but i also want to know

864
00:34:07,100 --> 00:34:08,629
respect all of your time so i

865
00:34:08,633 --> 00:34:10,463
love ending early and i have more

866
00:34:10,466 --> 00:34:12,386
time tomorrow so if you have deep

867
00:34:12,366 --> 00:34:14,195
questions and want to work on this

868
00:34:14,199 --> 00:34:16,841
more with us tomorrow is a good

869
00:34:16,866 --> 00:34:17,496
time as well

870
00:34:19,433 --> 00:34:20,543
never said the coffee

871
00:34:32,766 --> 00:34:33,299
i

872
00:34:37,433 --> 00:34:38,694
yeah i think we're also i mean

873
00:34:39,033 --> 00:34:41,300
in an area that while i'm waiting

874
00:34:41,300 --> 00:34:44,060
for questions please ask away and we

875
00:34:44,066 --> 00:34:45,386
do want to talk to we know

876
00:34:45,366 --> 00:34:46,806
a lot of people use matomo with

877
00:34:46,800 --> 00:34:48,870
web analytics but we are really interested

878
00:34:48,866 --> 00:34:50,306
in people that are using them a

879
00:34:50,333 --> 00:34:53,752
tomo as mobile st case and and

880
00:34:53,800 --> 00:34:54,594
and how you've been

881
00:34:54,566 --> 00:34:56,546
degraded into native apps would be great

882
00:34:56,566 --> 00:34:59,506
to hear from you either now or

883
00:34:59,500 --> 00:35:01,150
tomorrow or through the chat rooms

884
00:35:02,666 --> 00:35:04,616
and to kind of talk about how

885
00:35:05,366 --> 00:35:06,626
you we might you know you can

886
00:35:06,633 --> 00:35:09,453
migrate to using clean insights the sdk

887
00:35:09,466 --> 00:35:10,486
instead perhaps

888
00:35:11,733 --> 00:35:12,903
or why you wouldn't

889
00:35:36,200 --> 00:35:38,839
thanks beyond our hey appreciate that

890
00:35:40,100 --> 00:35:42,379
goodness this easy softball i will say

891
00:35:42,366 --> 00:35:45,756
and boy are hans just posted something

892
00:35:45,766 --> 00:35:48,196
in our our feet about how android

893
00:35:48,200 --> 00:35:49,879
apps can track users based on their

894
00:35:49,866 --> 00:35:54,606
wallpapers and and it's you know there's

895
00:35:54,600 --> 00:35:56,430
so many problems on mobile devices in

896
00:35:56,433 --> 00:35:57,483
this area with

897
00:35:58,733 --> 00:36:01,883
you know perma cookies are essentially fingerprinting

898
00:36:01,866 --> 00:36:03,066
a device as a way to get

899
00:36:03,066 --> 00:36:06,786
a perma cookie and i guess wallpapers

900
00:36:06,800 --> 00:36:09,200
the latest so i i was

901
00:36:10,433 --> 00:36:13,218
you know i often set custom wallpapers

902
00:36:13,233 --> 00:36:14,918
so i'm you know that's a problem

903
00:36:14,933 --> 00:36:18,563
i have and yeah okay well thank

904
00:36:18,566 --> 00:36:21,296
you for all the kind words and

905
00:36:22,433 --> 00:36:24,923
i will be attending other presentations today

906
00:36:24,933 --> 00:36:26,643
as well so hope to hear all

907
00:36:26,666 --> 00:36:29,276
of your other work there and seeing

908
00:36:29,300 --> 00:36:30,260
i know there's a lot of other

909
00:36:30,266 --> 00:36:33,146
relevant topics so but again please come

910
00:36:33,133 --> 00:36:34,693
tomorrow morning if you want to talk

911
00:36:34,700 --> 00:36:37,669
through your own implementation or sign up

912
00:36:37,666 --> 00:36:39,976
at clean insights dot org slash beans

913
00:36:40,266 --> 00:36:42,576
for a private consultation with caf

914
00:36:42,566 --> 00:36:45,446
lee and stickers to get stickers there

915
00:36:45,466 --> 00:36:45,796
we go

916
00:36:48,200 --> 00:36:49,490
i think we're done then if there's

917
00:36:49,466 --> 00:36:50,306
no questions

918
00:36:54,866 --> 00:36:57,506
many thanks nita i think people are

919
00:36:57,500 --> 00:36:59,570
a little bit shy for the moment

920
00:36:59,800 --> 00:37:02,500
but it can also continue the conversation

921
00:37:02,500 --> 00:37:04,720
in the chant sure i'll be there

922
00:37:04,733 --> 00:37:06,113
and in the cafe as well

923
00:37:08,266 --> 00:37:09,796
thank you thank you

924
00:37:25,466 --> 00:37:27,056
you should i have my ukulele or

925
00:37:27,033 --> 00:37:28,203
guitar here and i can play a

926
00:37:28,200 --> 00:37:29,669
song but i can't find it at

927
00:37:29,666 --> 00:37:30,116
the moment

928
00:37:33,033 --> 00:37:33,453
hm

929
00:37:34,766 --> 00:37:38,486
won't be found though no next time

930
00:37:39,900 --> 00:37:42,930
know where it went children took it

931
00:37:47,133 --> 00:37:49,533
i have been home from a date

932
00:37:49,833 --> 00:37:51,543
the

933
00:37:58,200 --> 00:37:59,640
alright this is an

934
00:38:01,433 --> 00:38:04,283
this instrument is a i'm nathan by

935
00:38:04,266 --> 00:38:05,526
the way if everyone's wanted this is

936
00:38:05,533 --> 00:38:08,203
called a drum yen or tibetan banjo

937
00:38:09,133 --> 00:38:11,173
and it has double strings

938
00:38:14,166 --> 00:38:16,086
and the interesting thing is the bottom

939
00:38:16,100 --> 00:38:17,330
string is kind of a drone

940
00:38:18,366 --> 00:38:19,086
like a bass

941
00:38:21,333 --> 00:38:22,653
and it's not tuned up because this

942
00:38:22,666 --> 00:38:25,126
is more of an instrumental ornamental instrument

943
00:38:25,133 --> 00:38:27,203
but yeah it's a really beautiful instrument

944
00:38:27,266 --> 00:38:28,946
and er

945
00:38:38,666 --> 00:38:40,706
this is just waiting music or figuring

946
00:38:40,700 --> 00:38:42,890
out the technical difficulties anyway

947
00:38:43,666 --> 00:38:45,136
it's a pretty fun instrument it's got

948
00:38:45,133 --> 00:38:47,353
this cool like bass drone like like

949
00:38:47,366 --> 00:38:49,436
i'm kind of tibetan mongolian you know

950
00:38:49,433 --> 00:38:51,351
that throat singing sound that was a

951
00:38:51,333 --> 00:38:52,685
real low base and then you play

952
00:38:52,700 --> 00:38:55,249
the melody on the top strings and

953
00:38:55,266 --> 00:38:57,456
yeah i brought this back from tibet

954
00:38:57,466 --> 00:38:59,416
many years ago this beautiful skin there

955
00:38:59,900 --> 00:39:00,740
okay we're going to go with the

956
00:39:00,733 --> 00:39:01,633
pdf file