0 00:00:00,000 --> 00:00:30,000 Dear viewer, these subtitles were generated by a machine via the service Trint and therefore are (very) buggy. If you are capable, please help us to create good quality subtitles: https://c3subtitles.de/talk/888 Thanks! 1 00:00:15,680 --> 00:00:17,609 Welcome, everybody, to our next talk 2 00:00:17,610 --> 00:00:19,729 only, let's fight Internet censorship 3 00:00:19,730 --> 00:00:21,049 together. 4 00:00:21,050 --> 00:00:23,329 When you think about Internet censorship, 5 00:00:23,330 --> 00:00:25,669 I guess that a lot of people would just 6 00:00:25,670 --> 00:00:27,829 have an association with countries like 7 00:00:27,830 --> 00:00:29,899 China or Iran or Saudi 8 00:00:29,900 --> 00:00:31,069 Arabia. 9 00:00:31,070 --> 00:00:32,959 But not a lot of people are aware that 10 00:00:32,960 --> 00:00:35,059 filtering and blocking of websites 11 00:00:35,060 --> 00:00:37,069 and Internet censorship also takes place 12 00:00:37,070 --> 00:00:39,079 and almost all the other countries around 13 00:00:39,080 --> 00:00:39,979 the world. 14 00:00:39,980 --> 00:00:42,289 And this talk will educate you about 15 00:00:42,290 --> 00:00:44,389 this. Our speaker is 16 00:00:44,390 --> 00:00:46,459 to feel as though he is the 17 00:00:46,460 --> 00:00:48,799 founder and developer of UNY, 18 00:00:48,800 --> 00:00:50,299 and he will tell you more about this 19 00:00:50,300 --> 00:00:52,429 super interesting project of his in this 20 00:00:52,430 --> 00:00:54,349 talk. Please give him a warm round of 21 00:00:54,350 --> 00:00:55,350 applause. 22 00:00:57,330 --> 00:00:58,330 Thank you, Miriam. 23 00:01:00,170 --> 00:01:02,809 Hi, erm, 24 00:01:02,810 --> 00:01:04,759 I guess I would like to start with 25 00:01:04,760 --> 00:01:07,159 thanking you and sort of 26 00:01:07,160 --> 00:01:09,289 expressing how much this is an honor 27 00:01:09,290 --> 00:01:11,689 for me to present this project 28 00:01:11,690 --> 00:01:14,599 in front of an audience like C.C.C.. 29 00:01:14,600 --> 00:01:16,879 I've been coming to to these events 30 00:01:16,880 --> 00:01:19,189 since I was just just 31 00:01:19,190 --> 00:01:21,499 over 18 years of age. 32 00:01:21,500 --> 00:01:24,449 And so I think this project 33 00:01:24,450 --> 00:01:26,539 and also the 34 00:01:26,540 --> 00:01:28,909 a large part of my adult life 35 00:01:28,910 --> 00:01:31,039 is also due to the encounters and 36 00:01:31,040 --> 00:01:33,229 the things I have learned at events like 37 00:01:33,230 --> 00:01:34,230 this. 38 00:01:35,030 --> 00:01:36,030 And so 39 00:01:37,160 --> 00:01:38,209 to get right into it, 40 00:01:39,500 --> 00:01:41,809 I think the first thing that 41 00:01:41,810 --> 00:01:44,269 we should start with when talking about 42 00:01:44,270 --> 00:01:46,550 a project like Erni is. 43 00:01:47,790 --> 00:01:50,849 Understand what we are talking about, so 44 00:01:50,850 --> 00:01:52,949 to frame what what we mean when we 45 00:01:52,950 --> 00:01:55,169 say Internet censorship, 46 00:01:56,220 --> 00:01:58,469 because these 47 00:01:58,470 --> 00:01:59,470 are these. 48 00:02:02,980 --> 00:02:05,229 There is some video 49 00:02:05,230 --> 00:02:06,230 connection. 50 00:02:07,030 --> 00:02:09,129 OK, yeah, so what 51 00:02:09,130 --> 00:02:10,269 is Internet censorship 52 00:02:12,010 --> 00:02:13,599 within the context of Sony? 53 00:02:13,600 --> 00:02:15,969 When we talk about Internet censorship, 54 00:02:15,970 --> 00:02:18,339 what we mean is some form 55 00:02:18,340 --> 00:02:20,889 of network interference 56 00:02:20,890 --> 00:02:23,919 that comes with intent 57 00:02:23,920 --> 00:02:26,949 to disrupt or distort 58 00:02:26,950 --> 00:02:29,199 the reality of 59 00:02:29,200 --> 00:02:31,479 of the Internet from the perspective 60 00:02:31,480 --> 00:02:33,180 of a country or a network. 61 00:02:34,210 --> 00:02:36,550 So it means that it requires 62 00:02:37,690 --> 00:02:40,449 intent from the censor to 63 00:02:40,450 --> 00:02:42,639 create this altered 64 00:02:42,640 --> 00:02:43,780 view of what is 65 00:02:45,460 --> 00:02:46,689 the Internet. 66 00:02:46,690 --> 00:02:49,419 Another term that we use to define 67 00:02:49,420 --> 00:02:52,179 Internet censorship is is a filter net. 68 00:02:52,180 --> 00:02:54,219 And what we will see in during the course 69 00:02:54,220 --> 00:02:55,899 of this presentation is that there are, 70 00:02:55,900 --> 00:02:57,969 in fact, many filter nets around the 71 00:02:57,970 --> 00:03:00,819 world and that depending on 72 00:03:00,820 --> 00:03:03,249 which country you are in and depending 73 00:03:03,250 --> 00:03:05,439 on the on the regime 74 00:03:05,440 --> 00:03:07,659 or the ideology 75 00:03:07,660 --> 00:03:09,939 of those in power, 76 00:03:09,940 --> 00:03:12,069 different views of 77 00:03:12,070 --> 00:03:14,109 of what is acceptable and not acceptable 78 00:03:14,110 --> 00:03:16,479 to be done on the Internet 79 00:03:16,480 --> 00:03:17,679 are enforced 80 00:03:19,930 --> 00:03:21,999 and so on. 81 00:03:22,000 --> 00:03:23,679 Or the open observatory of network 82 00:03:23,680 --> 00:03:25,779 interference is this 83 00:03:25,780 --> 00:03:27,579 network measurement project that was 84 00:03:27,580 --> 00:03:29,919 started back in 2012. 85 00:03:29,920 --> 00:03:32,049 And it's a community 86 00:03:32,050 --> 00:03:34,330 led project 87 00:03:35,440 --> 00:03:36,969 that's composed of a network of 88 00:03:36,970 --> 00:03:39,189 volunteers that install 89 00:03:39,190 --> 00:03:41,409 the software and collect 90 00:03:41,410 --> 00:03:43,689 evidence of Internet censorship around 91 00:03:43,690 --> 00:03:45,580 the world since 92 00:03:47,140 --> 00:03:49,839 until today we have gathered 93 00:03:49,840 --> 00:03:52,839 measurements from over 200 countries 94 00:03:52,840 --> 00:03:55,149 and we have measurements 95 00:03:55,150 --> 00:03:57,519 coming in from tens of thousands 96 00:03:57,520 --> 00:03:59,229 of different networks every month. 97 00:04:01,450 --> 00:04:02,450 But 98 00:04:03,550 --> 00:04:06,639 the sort of founding principles 99 00:04:06,640 --> 00:04:08,799 around uni and the reason why we 100 00:04:08,800 --> 00:04:10,929 decided to work on this 101 00:04:10,930 --> 00:04:12,999 project is that we 102 00:04:14,190 --> 00:04:16,299 we want to sort of bring to 103 00:04:16,300 --> 00:04:17,799 the field of network measurement and in 104 00:04:17,800 --> 00:04:19,789 particular the field of network 105 00:04:19,790 --> 00:04:22,779 measurement applied to detecting 106 00:04:22,780 --> 00:04:25,029 network interference and 107 00:04:25,030 --> 00:04:27,399 or Internet censorship, the principles 108 00:04:27,400 --> 00:04:29,469 of openness and transparency. 109 00:04:29,470 --> 00:04:32,889 So only as open methodologies, 110 00:04:32,890 --> 00:04:35,289 which means that all of the tools 111 00:04:35,290 --> 00:04:38,409 and techniques that we use in order to 112 00:04:38,410 --> 00:04:40,869 examine Internet censorship are 113 00:04:40,870 --> 00:04:43,599 openly documented and specified, 114 00:04:43,600 --> 00:04:45,939 sort of like in RACC format. 115 00:04:45,940 --> 00:04:48,549 And and these allow 116 00:04:48,550 --> 00:04:51,069 people to validate 117 00:04:51,070 --> 00:04:53,559 that. What we are claiming to do is 118 00:04:53,560 --> 00:04:55,659 actually reasonable and 119 00:04:55,660 --> 00:04:56,660 and fair. 120 00:04:58,180 --> 00:04:59,980 The methodologies that 121 00:05:01,540 --> 00:05:04,089 that we that we specify 122 00:05:04,090 --> 00:05:06,429 are then implemented using open 123 00:05:06,430 --> 00:05:07,869 source software. 124 00:05:07,870 --> 00:05:10,419 So you can verify that what we define 125 00:05:10,420 --> 00:05:12,189 in the specification is in fact 126 00:05:12,190 --> 00:05:14,739 consistent with what is implemented. 127 00:05:14,740 --> 00:05:16,929 And all the data that is collected 128 00:05:16,930 --> 00:05:18,999 by the thousands and 129 00:05:19,000 --> 00:05:20,259 tens of thousands and hundreds of 130 00:05:20,260 --> 00:05:22,749 thousands of volunteers around the world 131 00:05:22,750 --> 00:05:24,969 are made available openly as open data. 132 00:05:24,970 --> 00:05:27,129 So anybody can use this data 133 00:05:27,130 --> 00:05:29,229 for their own purpose and sort 134 00:05:29,230 --> 00:05:31,389 of from from the beginning to the end 135 00:05:31,390 --> 00:05:33,699 of the chain, validate that 136 00:05:33,700 --> 00:05:36,219 what we are doing is, in fact 137 00:05:36,220 --> 00:05:38,809 correct and and reasonable. 138 00:05:38,810 --> 00:05:40,869 And hopefully by 139 00:05:40,870 --> 00:05:42,939 doing all of this, we will reach at 140 00:05:42,940 --> 00:05:45,099 some point a state, 141 00:05:45,100 --> 00:05:47,259 a stage in which we can 142 00:05:47,260 --> 00:05:49,089 we can say that what we are doing is, in 143 00:05:49,090 --> 00:05:51,489 fact, reproducible research 144 00:05:51,490 --> 00:05:53,919 so that every step 145 00:05:53,920 --> 00:05:56,079 from the definition of what you do to 146 00:05:56,080 --> 00:05:57,909 the implementation, to the publication of 147 00:05:57,910 --> 00:06:00,459 the data is entirely reproducible. 148 00:06:00,460 --> 00:06:02,529 All of our tooling, like all 149 00:06:02,530 --> 00:06:04,149 of our back in infrastructure, even our 150 00:06:04,150 --> 00:06:05,799 sysadmin scripts, are made available 151 00:06:05,800 --> 00:06:07,959 openly so anybody can set up 152 00:06:07,960 --> 00:06:10,509 their own money if 153 00:06:10,510 --> 00:06:11,510 they wish. 154 00:06:12,700 --> 00:06:14,949 But before going more into what UNE 155 00:06:14,950 --> 00:06:17,139 is, I would like to to 156 00:06:17,140 --> 00:06:19,329 give you a taste 157 00:06:19,330 --> 00:06:21,609 of what what is possible 158 00:06:21,610 --> 00:06:23,829 with this sort of research, as in what 159 00:06:23,830 --> 00:06:25,899 are the sorts of things that can be 160 00:06:25,900 --> 00:06:28,629 discovered by collecting 161 00:06:28,630 --> 00:06:30,759 network measurements around 162 00:06:30,760 --> 00:06:32,049 the world. 163 00:06:32,050 --> 00:06:34,329 And these are obviously 164 00:06:34,330 --> 00:06:36,939 just some small samples. 165 00:06:36,940 --> 00:06:38,919 You can learn more about each of these 166 00:06:38,920 --> 00:06:41,259 case by reaching our full length reports. 167 00:06:41,260 --> 00:06:43,599 But this is sort of more to show 168 00:06:43,600 --> 00:06:45,789 you. What is the kind of impact 169 00:06:45,790 --> 00:06:47,859 that a tool like 170 00:06:47,860 --> 00:06:49,420 Huni probe can have in 171 00:06:50,710 --> 00:06:52,089 in the world? 172 00:06:52,090 --> 00:06:53,649 So the first case I would like to talk to 173 00:06:53,650 --> 00:06:55,809 you about is one that 174 00:06:55,810 --> 00:06:57,999 maybe you have heard in the news, or 175 00:06:58,000 --> 00:07:00,099 maybe not, because it was not covered as 176 00:07:00,100 --> 00:07:01,689 well. And it's. 177 00:07:01,690 --> 00:07:04,359 Round the the protests in Ethiopia 178 00:07:04,360 --> 00:07:06,609 in December of last 179 00:07:06,610 --> 00:07:07,740 year, 2016, 180 00:07:09,130 --> 00:07:11,229 these protests sparked in a 181 00:07:11,230 --> 00:07:13,029 particular region region of the GOP 182 00:07:13,030 --> 00:07:15,099 called the Oromo region, and they 183 00:07:15,100 --> 00:07:17,439 were around some conflict related 184 00:07:17,440 --> 00:07:19,689 to the expansion of 185 00:07:19,690 --> 00:07:21,999 a certain city of 186 00:07:22,000 --> 00:07:24,120 the borders of the city, Addis Ababa. 187 00:07:25,510 --> 00:07:27,459 That would, in a way, 188 00:07:30,010 --> 00:07:32,319 sort of basically kick people 189 00:07:32,320 --> 00:07:34,119 out of their homes because they would 190 00:07:34,120 --> 00:07:36,549 suddenly have a big house 191 00:07:36,550 --> 00:07:39,789 like a big city in where they live. 192 00:07:39,790 --> 00:07:42,129 And so this this led to some very 193 00:07:42,130 --> 00:07:44,439 serious protests that 194 00:07:44,440 --> 00:07:45,459 were very violent. 195 00:07:45,460 --> 00:07:48,639 And the police was also very 196 00:07:48,640 --> 00:07:51,159 brutal to the to the protesters. 197 00:07:51,160 --> 00:07:52,389 And it led to many deaths. 198 00:07:54,690 --> 00:07:56,829 And and as a result of these 199 00:07:56,830 --> 00:07:59,139 protests, also some 200 00:07:59,140 --> 00:08:00,789 network interference was done. 201 00:08:00,790 --> 00:08:03,609 So in order to stop 202 00:08:03,610 --> 00:08:05,349 the protesters from communicating and 203 00:08:05,350 --> 00:08:08,109 organizing, WhatsApp was blocked 204 00:08:08,110 --> 00:08:10,419 so the protesters could no longer 205 00:08:10,420 --> 00:08:12,789 coordinate or organize 206 00:08:12,790 --> 00:08:13,809 themselves. 207 00:08:13,810 --> 00:08:15,639 And with unity, we gathered evidence of 208 00:08:15,640 --> 00:08:17,649 this so we could prove in an undeniable 209 00:08:17,650 --> 00:08:20,619 way that WhatsApp was blocked, 210 00:08:20,620 --> 00:08:22,449 that deep packet inspection technology 211 00:08:22,450 --> 00:08:23,799 was being used. 212 00:08:23,800 --> 00:08:26,289 And we were also able to understand 213 00:08:26,290 --> 00:08:28,569 precisely how it was working. 214 00:08:28,570 --> 00:08:31,179 That allowed us even to give some 215 00:08:31,180 --> 00:08:33,339 advice of what some site operators 216 00:08:33,340 --> 00:08:35,529 could do to even circumvented 217 00:08:35,530 --> 00:08:36,530 in some way. 218 00:08:37,510 --> 00:08:39,759 And many websites from 219 00:08:39,760 --> 00:08:41,949 media outlets, human rights 220 00:08:41,950 --> 00:08:43,359 websites and obviously political 221 00:08:43,360 --> 00:08:45,519 opposition websites were found 222 00:08:45,520 --> 00:08:46,990 to be blocked during this period 223 00:08:48,550 --> 00:08:49,509 following this. 224 00:08:49,510 --> 00:08:51,789 In Ethiopia, the state 225 00:08:51,790 --> 00:08:54,399 of a state of emergency was declared 226 00:08:54,400 --> 00:08:56,649 starting from October 8th, 227 00:08:56,650 --> 00:08:59,229 2016, just until 228 00:08:59,230 --> 00:09:01,479 a couple of some months ago, 229 00:09:01,480 --> 00:09:03,399 August 20, 17. 230 00:09:03,400 --> 00:09:05,799 So lasting 10 months. 231 00:09:05,800 --> 00:09:07,299 But what's interesting from this 232 00:09:08,590 --> 00:09:10,929 sort of case is that what we found 233 00:09:10,930 --> 00:09:13,359 evidence of is actually blocking 234 00:09:13,360 --> 00:09:15,669 and censorship being implemented before 235 00:09:15,670 --> 00:09:17,799 the state of emergency, 236 00:09:17,800 --> 00:09:19,719 which in theory was something that was 237 00:09:19,720 --> 00:09:21,789 not supposed to be allowed. 238 00:09:23,530 --> 00:09:26,019 And together with Amnesty International, 239 00:09:26,020 --> 00:09:28,449 we published a research report 240 00:09:28,450 --> 00:09:31,419 outlining our findings and 241 00:09:31,420 --> 00:09:33,699 sort of providing evidence 242 00:09:33,700 --> 00:09:35,859 that this was 243 00:09:35,860 --> 00:09:37,329 something that that was, in fact, 244 00:09:37,330 --> 00:09:39,489 happening and and was clearly 245 00:09:40,600 --> 00:09:41,710 not something that 246 00:09:43,150 --> 00:09:45,669 a clear human rights violation. 247 00:09:47,070 --> 00:09:48,070 Um. 248 00:09:49,330 --> 00:09:51,729 Another interesting case 249 00:09:51,730 --> 00:09:54,039 that we worked on together with our 250 00:09:54,040 --> 00:09:56,239 our partner, our Malaysian partner 251 00:09:56,240 --> 00:09:58,929 senior project was 252 00:09:58,930 --> 00:10:01,149 published in May 2017, and it was 253 00:10:01,150 --> 00:10:03,489 a sort of a study of Internet 254 00:10:03,490 --> 00:10:05,359 censorship in in Indonesia. 255 00:10:06,940 --> 00:10:07,940 We found 256 00:10:09,760 --> 00:10:11,139 that many, 257 00:10:12,640 --> 00:10:14,739 many sites of various categories 258 00:10:14,740 --> 00:10:15,999 were blocked. 259 00:10:16,000 --> 00:10:18,039 But I think one of the most interesting 260 00:10:18,040 --> 00:10:20,409 things from from this particular 261 00:10:20,410 --> 00:10:23,259 report was the fact that 262 00:10:23,260 --> 00:10:24,820 we collected measurements from 263 00:10:26,260 --> 00:10:28,959 around 60 different networks. 264 00:10:28,960 --> 00:10:30,609 And based on these, we were able to see 265 00:10:30,610 --> 00:10:32,919 that there was actually very much some 266 00:10:32,920 --> 00:10:34,509 difference between the different 267 00:10:34,510 --> 00:10:36,579 operators in terms of what was being 268 00:10:36,580 --> 00:10:38,799 blocked and what was not being blocked. 269 00:10:38,800 --> 00:10:40,809 So as an example, Vimeo and Reddit 270 00:10:41,830 --> 00:10:43,629 were found to be blocked on certain 271 00:10:43,630 --> 00:10:46,299 Internet service providers in Indonesia, 272 00:10:46,300 --> 00:10:48,219 although the ban for blocking such 273 00:10:48,220 --> 00:10:50,409 websites was supposedly lifted two years 274 00:10:50,410 --> 00:10:51,410 ago. 275 00:10:52,630 --> 00:10:54,849 And this stems, I guess, from 276 00:10:54,850 --> 00:10:57,249 the fact that the Indonesian 277 00:10:57,250 --> 00:10:59,349 legislation states 278 00:10:59,350 --> 00:11:02,409 that Internet service providers 279 00:11:02,410 --> 00:11:05,319 are are granted the authority 280 00:11:05,320 --> 00:11:08,079 to inhibit or block content 281 00:11:08,080 --> 00:11:11,259 that is considered negative content 282 00:11:11,260 --> 00:11:13,599 and they are left 283 00:11:13,600 --> 00:11:16,089 on their own to make this judgment 284 00:11:16,090 --> 00:11:18,339 call of what constitutes negative 285 00:11:18,340 --> 00:11:19,509 content. 286 00:11:19,510 --> 00:11:20,649 And as a result, you see 287 00:11:21,970 --> 00:11:24,099 quite some variation in terms of 288 00:11:24,100 --> 00:11:25,749 which sites are blocked and which are not 289 00:11:25,750 --> 00:11:26,750 blocked. 290 00:11:27,820 --> 00:11:29,949 And this is why it's important even 291 00:11:29,950 --> 00:11:32,139 in places where you may say, you 292 00:11:32,140 --> 00:11:34,269 know, that censorship is happening. 293 00:11:34,270 --> 00:11:35,769 Yes, you may know that censorship is 294 00:11:35,770 --> 00:11:38,439 happening, but you don't necessarily know 295 00:11:38,440 --> 00:11:40,389 what the censorship of your neighbor 296 00:11:40,390 --> 00:11:42,519 looks like, even if they live in 297 00:11:42,520 --> 00:11:43,520 the same country. 298 00:11:46,340 --> 00:11:47,340 Another 299 00:11:48,500 --> 00:11:50,689 very interesting report that 300 00:11:50,690 --> 00:11:53,569 I think we could spend an entire 301 00:11:53,570 --> 00:11:56,029 one hour talk just 302 00:11:56,030 --> 00:11:58,549 discussing about this 303 00:11:58,550 --> 00:11:59,550 particular country 304 00:12:00,770 --> 00:12:03,979 was published in September 2017 305 00:12:03,980 --> 00:12:06,109 and it was on the country 306 00:12:06,110 --> 00:12:07,669 of Iran. 307 00:12:07,670 --> 00:12:10,789 And there we found an astounding 308 00:12:10,790 --> 00:12:13,009 number of of websites 309 00:12:13,010 --> 00:12:14,599 to be to be blocked. 310 00:12:14,600 --> 00:12:16,729 I think in in total, we 311 00:12:16,730 --> 00:12:17,929 counted one thousand 312 00:12:19,100 --> 00:12:20,929 and 20 year olds 313 00:12:22,190 --> 00:12:25,039 blocked, ranging from 314 00:12:25,040 --> 00:12:26,899 all sorts of various different 315 00:12:26,900 --> 00:12:28,489 categories. 316 00:12:28,490 --> 00:12:30,589 And without going too much in depth into 317 00:12:30,590 --> 00:12:32,689 this, what I 318 00:12:32,690 --> 00:12:34,729 should say on Iran is that it's 319 00:12:34,730 --> 00:12:37,189 definitely one of the countries that has 320 00:12:37,190 --> 00:12:39,589 the most state of the art Internet 321 00:12:39,590 --> 00:12:41,779 censorship apparatuses, and 322 00:12:41,780 --> 00:12:42,860 they implement the most 323 00:12:44,480 --> 00:12:46,519 sort of advanced forms of Internet 324 00:12:46,520 --> 00:12:47,520 censorship. 325 00:12:48,140 --> 00:12:50,209 And two interesting aspects that 326 00:12:50,210 --> 00:12:52,429 are unique to this country, I think 327 00:12:52,430 --> 00:12:54,709 are, one, that the censorship 328 00:12:54,710 --> 00:12:57,199 that they implement is done 329 00:12:57,200 --> 00:12:59,719 in, let's say, nondeterministic 330 00:12:59,720 --> 00:13:00,829 fashion. 331 00:13:00,830 --> 00:13:02,449 What this means is that you will visit a 332 00:13:02,450 --> 00:13:04,309 website, and it's not that every 333 00:13:04,310 --> 00:13:05,809 connection to that website will be 334 00:13:05,810 --> 00:13:07,939 terminated immediately 335 00:13:07,940 --> 00:13:09,139 or every time you visit it. 336 00:13:09,140 --> 00:13:11,479 They will just selectively drop 337 00:13:11,480 --> 00:13:13,789 some packets, every certain 338 00:13:13,790 --> 00:13:15,949 number of requests leading 339 00:13:15,950 --> 00:13:19,039 the user to think in some cases that 340 00:13:19,040 --> 00:13:21,109 the website is not, in fact blocked, but 341 00:13:21,110 --> 00:13:23,359 is just unreliable or not 342 00:13:23,360 --> 00:13:24,750 not functioning properly. 343 00:13:26,780 --> 00:13:29,029 The other aspect that we found 344 00:13:29,030 --> 00:13:32,029 that was quite interesting is 345 00:13:32,030 --> 00:13:33,469 given the fact that we analyzed the 346 00:13:33,470 --> 00:13:36,499 corpus of data dating back to 2014, 347 00:13:36,500 --> 00:13:38,749 we were also able to track the evolution 348 00:13:38,750 --> 00:13:40,639 of censorship over the course of these 349 00:13:40,640 --> 00:13:41,769 years. 350 00:13:41,770 --> 00:13:43,969 And one of the things that we found 351 00:13:43,970 --> 00:13:46,819 was that certain websites 352 00:13:46,820 --> 00:13:49,069 such as Instagram, Instagram, 353 00:13:49,070 --> 00:13:51,049 that previously they were blocking only 354 00:13:51,050 --> 00:13:52,189 specific posts. 355 00:13:52,190 --> 00:13:55,069 So specific images or specific 356 00:13:55,070 --> 00:13:57,080 pages on the site that were deemed 357 00:13:58,250 --> 00:14:00,679 sort of unacceptable or 358 00:14:00,680 --> 00:14:02,929 against the, 359 00:14:02,930 --> 00:14:05,449 you know, the the morale of 360 00:14:05,450 --> 00:14:06,450 the majority. 361 00:14:07,520 --> 00:14:09,859 But as Instagram rolled 362 00:14:09,860 --> 00:14:13,099 out HTP on all of their websites, 363 00:14:13,100 --> 00:14:14,749 the Iranian government was put in a 364 00:14:14,750 --> 00:14:16,849 situation where they either blocked 365 00:14:16,850 --> 00:14:18,979 all the website or 366 00:14:18,980 --> 00:14:21,769 they could not block the specific pages. 367 00:14:21,770 --> 00:14:23,899 And so, I mean, I guess this is more 368 00:14:23,900 --> 00:14:27,259 something to reflect upon of how 369 00:14:27,260 --> 00:14:30,289 sometimes something that you take as just 370 00:14:30,290 --> 00:14:31,939 a positive thing. And it clearly is a 371 00:14:31,940 --> 00:14:34,729 positive thing to have websites 372 00:14:34,730 --> 00:14:37,189 implement and have 373 00:14:37,190 --> 00:14:39,079 secure connections to them. 374 00:14:39,080 --> 00:14:41,539 But it does also lead sometimes to some 375 00:14:41,540 --> 00:14:43,999 some effects that are not 376 00:14:44,000 --> 00:14:45,230 not expected maybe. 377 00:14:46,280 --> 00:14:47,449 And this is an example of it. 378 00:14:47,450 --> 00:14:50,419 So now Instagram is entirely blocked 379 00:14:50,420 --> 00:14:52,700 in in certain networks in Iran. 380 00:14:55,980 --> 00:14:58,529 Another case that probably 381 00:14:58,530 --> 00:14:59,530 you have heard about 382 00:15:01,020 --> 00:15:03,449 happened in in Spain 383 00:15:03,450 --> 00:15:05,939 just a couple of months ago, 384 00:15:05,940 --> 00:15:08,909 and that was around the 385 00:15:08,910 --> 00:15:12,039 Catalonia independence referendum. 386 00:15:12,040 --> 00:15:14,169 And there 387 00:15:14,170 --> 00:15:16,359 was some movement 388 00:15:16,360 --> 00:15:18,669 to do a referendum 389 00:15:18,670 --> 00:15:21,039 to claim independence from Spain 390 00:15:21,040 --> 00:15:24,009 for the Spanish region, Catalonia, 391 00:15:24,010 --> 00:15:26,139 and as a result of this, 392 00:15:26,140 --> 00:15:28,389 the Spanish government decided 393 00:15:28,390 --> 00:15:31,089 that one one way to 394 00:15:31,090 --> 00:15:33,729 sort of create an obstacle 395 00:15:33,730 --> 00:15:36,579 for the success of this, 396 00:15:36,580 --> 00:15:38,809 this movement was that of 397 00:15:38,810 --> 00:15:40,569 of implementing censorship and blocking 398 00:15:40,570 --> 00:15:41,570 websites. 399 00:15:43,000 --> 00:15:45,129 I think the reason why this is 400 00:15:45,130 --> 00:15:47,979 something important to talk about 401 00:15:47,980 --> 00:15:49,929 here in Europe is that we should not 402 00:15:49,930 --> 00:15:52,179 forget that these things also 403 00:15:52,180 --> 00:15:54,519 happen in our home and that oftentimes 404 00:15:54,520 --> 00:15:56,859 when we talk or think about Internet 405 00:15:56,860 --> 00:15:59,049 censorship, we think as 406 00:15:59,050 --> 00:16:01,419 it's something that happens only 407 00:16:01,420 --> 00:16:02,799 over there. 408 00:16:02,800 --> 00:16:05,019 But in fact, many countries in Europe 409 00:16:05,020 --> 00:16:07,149 implement censorship for for gambling 410 00:16:07,150 --> 00:16:08,619 or other reasons. 411 00:16:08,620 --> 00:16:10,149 And it's kind of a slippery slope, 412 00:16:10,150 --> 00:16:12,309 because once you have the infrastructure 413 00:16:12,310 --> 00:16:14,729 in place that can be used 414 00:16:14,730 --> 00:16:17,049 to implement Internet censorship, 415 00:16:17,050 --> 00:16:18,050 those in power 416 00:16:19,330 --> 00:16:22,089 can possibly abuse it to, 417 00:16:22,090 --> 00:16:24,159 you know, do things that 418 00:16:24,160 --> 00:16:26,440 are clearly not acceptable. 419 00:16:27,520 --> 00:16:29,469 And so during the Catalan referendum, 420 00:16:29,470 --> 00:16:32,169 twenty five websites were blocked. 421 00:16:32,170 --> 00:16:34,329 The DOT cat registry 422 00:16:34,330 --> 00:16:36,459 was raided by the police 423 00:16:36,460 --> 00:16:38,559 and they forced them to 424 00:16:38,560 --> 00:16:40,629 take down or block a 425 00:16:40,630 --> 00:16:43,659 series of dot cat domains 426 00:16:43,660 --> 00:16:46,569 that were hosting just information 427 00:16:46,570 --> 00:16:49,149 about the Catalonia independence 428 00:16:49,150 --> 00:16:50,150 referendum. 429 00:16:52,560 --> 00:16:54,779 So I hope this has given 430 00:16:54,780 --> 00:16:56,939 you an idea of what are the 431 00:16:56,940 --> 00:16:59,010 sort of things that can come out of 432 00:17:00,660 --> 00:17:02,999 running a software like probe and 433 00:17:03,000 --> 00:17:05,249 collecting this sort of evidence. 434 00:17:05,250 --> 00:17:07,379 So not what I would like to talk to 435 00:17:07,380 --> 00:17:09,449 you a bit about is the sort 436 00:17:09,450 --> 00:17:11,729 of ecosystem of 437 00:17:11,730 --> 00:17:14,189 Yuni software. So the various software 438 00:17:14,190 --> 00:17:16,379 components that make up Huni 439 00:17:16,380 --> 00:17:18,598 probe and all of the things that you 440 00:17:18,599 --> 00:17:20,338 see here are open source. 441 00:17:20,339 --> 00:17:22,559 You can like, you 442 00:17:22,560 --> 00:17:23,560 know. 443 00:17:24,420 --> 00:17:26,489 Git clone them and hack on 444 00:17:26,490 --> 00:17:27,490 them to 445 00:17:28,569 --> 00:17:29,569 to your pleasing 446 00:17:31,200 --> 00:17:33,569 end, and 447 00:17:33,570 --> 00:17:35,879 the first thing that I'll 448 00:17:35,880 --> 00:17:38,069 tell you a bit about is, is the sort of 449 00:17:38,070 --> 00:17:40,469 engine that uni probe, 450 00:17:40,470 --> 00:17:42,479 the client that we use to run the network 451 00:17:42,480 --> 00:17:44,999 experiments is is 452 00:17:45,000 --> 00:17:47,189 based on and it's called measurement kit. 453 00:17:47,190 --> 00:17:49,889 And this is a C++ library 454 00:17:49,890 --> 00:17:52,409 that we use to implement 455 00:17:52,410 --> 00:17:53,819 all of our network measurement 456 00:17:53,820 --> 00:17:55,329 experiments. 457 00:17:55,330 --> 00:17:57,749 And it's currently supports 458 00:17:57,750 --> 00:18:00,839 Android iOS as our two primary targets. 459 00:18:00,840 --> 00:18:03,479 But it also works on any Unix system 460 00:18:03,480 --> 00:18:04,480 and we are 461 00:18:05,850 --> 00:18:08,249 in the process of 462 00:18:08,250 --> 00:18:09,899 ensuring that it works properly on 463 00:18:09,900 --> 00:18:11,549 Windows as well. 464 00:18:11,550 --> 00:18:13,969 We have JNI and no gigas bindings, 465 00:18:13,970 --> 00:18:16,169 so you can include this library 466 00:18:16,170 --> 00:18:18,689 inside of a project of yours 467 00:18:18,690 --> 00:18:20,939 where you need to do some particular 468 00:18:20,940 --> 00:18:23,759 experiments that involve 469 00:18:23,760 --> 00:18:24,760 measuring networks. 470 00:18:27,900 --> 00:18:30,179 As I said, the tool that we use 471 00:18:30,180 --> 00:18:32,819 to collect network measurement 472 00:18:32,820 --> 00:18:34,829 data is goony probe. 473 00:18:34,830 --> 00:18:37,049 We have an Android and iOS app 474 00:18:37,050 --> 00:18:38,699 that you can download on your on your 475 00:18:38,700 --> 00:18:39,599 phone. 476 00:18:39,600 --> 00:18:41,219 It's also available on Android. 477 00:18:42,300 --> 00:18:44,459 We have Mac OS 478 00:18:44,460 --> 00:18:47,549 and Linux as well 479 00:18:47,550 --> 00:18:49,979 in the form of Web UI. 480 00:18:49,980 --> 00:18:52,049 So but 481 00:18:52,050 --> 00:18:54,269 we are also working on a 482 00:18:54,270 --> 00:18:56,399 more fully feature, the proper desktop 483 00:18:56,400 --> 00:18:58,499 client that will support 484 00:18:58,500 --> 00:19:00,659 also also Windows as well as Mac 485 00:19:00,660 --> 00:19:01,660 and Linux. 486 00:19:03,660 --> 00:19:06,779 And the sort of tests that we do, 487 00:19:06,780 --> 00:19:09,179 as I guess you 488 00:19:09,180 --> 00:19:11,459 somewhat understood from from the 489 00:19:11,460 --> 00:19:14,099 the the case studies 490 00:19:14,100 --> 00:19:16,199 are mainly around these sorts 491 00:19:16,200 --> 00:19:17,549 of categories. 492 00:19:17,550 --> 00:19:19,919 So we have tests that examine 493 00:19:19,920 --> 00:19:20,849 Web censorship. 494 00:19:20,850 --> 00:19:22,799 So whether or not a particular website is 495 00:19:22,800 --> 00:19:24,929 blocked or is not blocked, 496 00:19:24,930 --> 00:19:26,849 we check to see if instant messaging apps 497 00:19:26,850 --> 00:19:29,189 like Facebook, Messenger, 498 00:19:29,190 --> 00:19:31,589 Telegram, WhatsApp, etc., 499 00:19:31,590 --> 00:19:33,329 work or do not work. 500 00:19:33,330 --> 00:19:35,129 We have a series of tests for censorship, 501 00:19:35,130 --> 00:19:37,649 circumvention tools like Tor, 502 00:19:37,650 --> 00:19:39,719 VPN, Lantern's 503 00:19:39,720 --> 00:19:40,720 iPhone, etc. 504 00:19:42,420 --> 00:19:44,519 Then we have some tests that sort 505 00:19:44,520 --> 00:19:47,219 of are more trying to understand 506 00:19:47,220 --> 00:19:48,749 certain characteristics about your 507 00:19:48,750 --> 00:19:51,989 network. So whether or not a middle box, 508 00:19:51,990 --> 00:19:54,569 so something like a transparent HGV proxy 509 00:19:54,570 --> 00:19:56,969 or something that is intercepting 510 00:19:56,970 --> 00:19:59,279 your connections and forwarding 511 00:19:59,280 --> 00:20:01,469 on them for you is present 512 00:20:01,470 --> 00:20:02,369 on your network. 513 00:20:02,370 --> 00:20:04,109 And more recently, we've started to have 514 00:20:04,110 --> 00:20:06,149 also tests for examining speed and 515 00:20:06,150 --> 00:20:07,150 performance. 516 00:20:08,400 --> 00:20:10,499 This is how our our mobile app 517 00:20:10,500 --> 00:20:12,659 looks like you can run it on 518 00:20:12,660 --> 00:20:15,029 your on your phone and and check 519 00:20:15,030 --> 00:20:17,159 to see what is blocked or not 520 00:20:17,160 --> 00:20:18,689 blocked on your network. 521 00:20:18,690 --> 00:20:21,299 And by doing so, you're contributing 522 00:20:21,300 --> 00:20:23,429 to our global understanding 523 00:20:23,430 --> 00:20:25,359 of Internet censorship around the world. 524 00:20:26,940 --> 00:20:28,949 This is our desktop clients that, as I 525 00:20:28,950 --> 00:20:30,960 was saying, is a Web user interface. 526 00:20:34,420 --> 00:20:36,579 And this is another interesting 527 00:20:36,580 --> 00:20:38,769 thing that we have recently launched, 528 00:20:38,770 --> 00:20:41,109 which is sort of a way for 529 00:20:41,110 --> 00:20:43,599 you to also engage other people 530 00:20:43,600 --> 00:20:44,999 to. 531 00:20:45,000 --> 00:20:47,159 Monitor or stand 532 00:20:47,160 --> 00:20:50,849 up for against Internet censorship, 533 00:20:50,850 --> 00:20:52,559 for the things that you care about. 534 00:20:52,560 --> 00:20:54,749 So with Iran, what you can do is you can 535 00:20:54,750 --> 00:20:56,909 add a list of your URLs that you want 536 00:20:56,910 --> 00:20:57,839 to monitor. 537 00:20:57,840 --> 00:20:59,999 So, for example, let's 538 00:21:00,000 --> 00:21:02,189 say animal 539 00:21:02,190 --> 00:21:04,619 rights activist 540 00:21:04,620 --> 00:21:06,629 group in Germany wants to monitor some 541 00:21:06,630 --> 00:21:07,649 German websites. 542 00:21:07,650 --> 00:21:09,779 They will create an yuni run 543 00:21:09,780 --> 00:21:10,979 link. 544 00:21:10,980 --> 00:21:12,899 And when people click on this, if they 545 00:21:12,900 --> 00:21:15,239 have the app installed, the app will open 546 00:21:15,240 --> 00:21:17,429 and will offer to test the sites 547 00:21:17,430 --> 00:21:19,109 that they want to test. 548 00:21:19,110 --> 00:21:21,419 So we obviously 549 00:21:21,420 --> 00:21:23,999 also accept contributions for 550 00:21:24,000 --> 00:21:26,099 or suggestions on what we we 551 00:21:26,100 --> 00:21:27,100 should be testing. 552 00:21:29,250 --> 00:21:30,250 But 553 00:21:31,860 --> 00:21:33,989 you can also do this on your own within 554 00:21:33,990 --> 00:21:34,990 Iran. 555 00:21:35,500 --> 00:21:37,379 And as I was saying, all the data is 556 00:21:37,380 --> 00:21:40,109 available as open data 557 00:21:40,110 --> 00:21:41,130 so you can 558 00:21:42,360 --> 00:21:44,579 access the API, visit the only 559 00:21:44,580 --> 00:21:46,690 explorer and 560 00:21:47,770 --> 00:21:49,949 and and use the data in your own 561 00:21:49,950 --> 00:21:52,299 studies to 562 00:21:52,300 --> 00:21:54,599 to uncover evidence of 563 00:21:54,600 --> 00:21:55,710 of Internet censorship. 564 00:21:57,750 --> 00:21:58,750 Sure. Why that is. 565 00:22:02,080 --> 00:22:04,149 So this is only Explorer. 566 00:22:04,150 --> 00:22:06,579 This is a global data map and it shows 567 00:22:06,580 --> 00:22:08,289 all the measurements that we have. 568 00:22:08,290 --> 00:22:11,079 You can browse through it to understand 569 00:22:11,080 --> 00:22:13,149 what is blocked and what measurements 570 00:22:13,150 --> 00:22:15,219 and what data is available 571 00:22:15,220 --> 00:22:16,599 in the various countries around the 572 00:22:16,600 --> 00:22:17,600 world. 573 00:22:18,310 --> 00:22:20,469 As I said, we also have an API 574 00:22:20,470 --> 00:22:22,509 that's more oriented towards, let's say, 575 00:22:22,510 --> 00:22:24,729 developers or data scientists. 576 00:22:24,730 --> 00:22:26,799 So you can query this API and 577 00:22:26,800 --> 00:22:29,079 obtain the raw data. 578 00:22:29,080 --> 00:22:30,819 And all the research reports that we 579 00:22:30,820 --> 00:22:33,369 write are actually all based on 580 00:22:33,370 --> 00:22:34,329 on the same data. 581 00:22:34,330 --> 00:22:36,639 So you can write your own research report 582 00:22:36,640 --> 00:22:38,589 on a country that you care about just by 583 00:22:38,590 --> 00:22:40,719 accessing our data, which is all, 584 00:22:40,720 --> 00:22:42,789 all, all as 585 00:22:42,790 --> 00:22:43,809 open as it can be. 586 00:22:45,190 --> 00:22:47,679 So get out there and 587 00:22:47,680 --> 00:22:50,259 use any data to understand 588 00:22:50,260 --> 00:22:52,449 better what is happening 589 00:22:52,450 --> 00:22:53,859 in the countries that you care about. 590 00:22:56,600 --> 00:22:58,759 And to close what I 591 00:22:58,760 --> 00:23:01,729 would like to leave you with is 592 00:23:01,730 --> 00:23:04,519 sort of another story, I guess, 593 00:23:04,520 --> 00:23:07,189 about a trip that 594 00:23:07,190 --> 00:23:09,619 some of the uni team did to Cuba. 595 00:23:11,060 --> 00:23:13,999 And I have a video here that 596 00:23:14,000 --> 00:23:16,789 I'll I'll use to support 597 00:23:16,790 --> 00:23:18,859 the narration of the story 598 00:23:18,860 --> 00:23:19,860 in a way. 599 00:23:20,810 --> 00:23:22,720 Um, and, um. 600 00:23:23,990 --> 00:23:26,059 And yeah. So this was in 601 00:23:26,060 --> 00:23:27,710 spring 2017 602 00:23:28,820 --> 00:23:29,820 and, um. 603 00:23:37,320 --> 00:23:39,389 Let's look and let's see if 604 00:23:39,390 --> 00:23:41,399 we can get this to start from the 605 00:23:41,400 --> 00:23:42,400 beginning. OK. 606 00:23:43,440 --> 00:23:45,749 OK, the audio is not playing. 607 00:23:45,750 --> 00:23:46,750 Yeah, so. 608 00:23:49,490 --> 00:23:51,769 Without further ado, I will tell you 609 00:23:51,770 --> 00:23:54,199 the story of our of our trip to Cuba 610 00:23:54,200 --> 00:23:56,779 and the data that you see here is 611 00:23:56,780 --> 00:23:58,999 and the facts here are obviously 612 00:23:59,000 --> 00:24:01,039 during the trip, some things have changed 613 00:24:01,040 --> 00:24:02,839 since we went there. 614 00:24:02,840 --> 00:24:03,840 But 615 00:24:05,690 --> 00:24:07,819 you can find more information 616 00:24:07,820 --> 00:24:09,829 inside of our of our research report. 617 00:24:14,180 --> 00:24:17,239 Many of us take the Internet for granted, 618 00:24:17,240 --> 00:24:18,799 many of the things here at C.C.C. 619 00:24:18,800 --> 00:24:20,390 would not be possible without it. 620 00:24:21,740 --> 00:24:24,319 But imagine living in a world 621 00:24:24,320 --> 00:24:26,419 where access to the Internet is not 622 00:24:26,420 --> 00:24:27,420 a big Quetta's. 623 00:24:28,180 --> 00:24:30,339 That's the type of world most Cubans live 624 00:24:30,340 --> 00:24:31,340 in. 625 00:24:31,600 --> 00:24:33,969 This is a story of our spring 626 00:24:33,970 --> 00:24:36,150 2017 trip to Cuba. 627 00:24:37,200 --> 00:24:38,880 Exploring its Internet landscape. 628 00:24:41,380 --> 00:24:43,300 Simona, Maria, 629 00:24:44,320 --> 00:24:46,479 Leonida, Joe and 630 00:24:46,480 --> 00:24:48,639 all of the other humanitarians that are 631 00:24:48,640 --> 00:24:51,669 here have 632 00:24:51,670 --> 00:24:53,359 this would not be possible without them. 633 00:24:53,360 --> 00:24:55,449 So this is also for for all of 634 00:24:55,450 --> 00:24:56,450 them. 635 00:24:57,070 --> 00:24:58,989 Most Cubans cannot access the Internet 636 00:24:58,990 --> 00:25:00,159 from the comfort of their homes. 637 00:25:00,160 --> 00:25:01,869 Rather, they need to visit designated 638 00:25:01,870 --> 00:25:04,779 public spaces that offer Internet access. 639 00:25:04,780 --> 00:25:07,359 Welcome to Park Net. 640 00:25:07,360 --> 00:25:09,279 This park, along with many others across 641 00:25:09,280 --> 00:25:11,079 the country, serves as a public Wi-Fi 642 00:25:11,080 --> 00:25:13,419 hotspot. Such hotspots were set up 643 00:25:13,420 --> 00:25:15,549 two years ago in 2015 when the Cuban 644 00:25:15,550 --> 00:25:17,379 government provided public access to the 645 00:25:17,380 --> 00:25:19,089 Internet for the first time. 646 00:25:19,090 --> 00:25:21,369 Most hotspots are located in parks, 647 00:25:21,370 --> 00:25:23,109 while others are located in other public 648 00:25:23,110 --> 00:25:26,079 spaces, like post offices and airports. 649 00:25:26,080 --> 00:25:28,269 In a way, Cubans don't access 650 00:25:28,270 --> 00:25:30,159 the Internet, they visit it. 651 00:25:30,160 --> 00:25:31,899 And so it becomes a sort of collective 652 00:25:31,900 --> 00:25:34,089 experience, especially 653 00:25:34,090 --> 00:25:36,669 at night when the heat is more tolerable. 654 00:25:36,670 --> 00:25:38,589 Dozens of Cubans gather in these hot 655 00:25:38,590 --> 00:25:40,749 spots to use the Internet and to 656 00:25:40,750 --> 00:25:42,039 chat online with their friends and 657 00:25:42,040 --> 00:25:43,040 family. 658 00:25:43,600 --> 00:25:45,549 To start browsing the net, you need to 659 00:25:45,550 --> 00:25:47,799 have an account issued by a taxi, 660 00:25:47,800 --> 00:25:49,839 Cuba's only telecommunications company. 661 00:25:54,130 --> 00:25:56,259 It Texas holds these cards at one point 662 00:25:56,260 --> 00:25:58,119 five dollars for one hour of Internet 663 00:25:58,120 --> 00:26:00,159 access. You can also purchase these 664 00:26:00,160 --> 00:26:02,589 accounts from resellers who 665 00:26:02,590 --> 00:26:04,479 lurk around in darkness, but they will 666 00:26:04,480 --> 00:26:06,489 mark up the price to even three or four 667 00:26:06,490 --> 00:26:08,619 dollars for only one hour of Internet 668 00:26:08,620 --> 00:26:09,620 access. 669 00:26:10,330 --> 00:26:12,399 We were curious to explore whether and 670 00:26:12,400 --> 00:26:14,529 to what extent Internet censorship was 671 00:26:14,530 --> 00:26:16,629 implemented in Cuba, and so 672 00:26:16,630 --> 00:26:18,189 we run our network measurement software 673 00:26:18,190 --> 00:26:19,929 called Zooni Probe, as well as a variety 674 00:26:19,930 --> 00:26:21,069 of other network tests. 675 00:26:22,210 --> 00:26:24,249 Based on our measurements, we found 41 676 00:26:24,250 --> 00:26:25,989 websites to be blocked. 677 00:26:25,990 --> 00:26:27,639 More might be blocked as we limit their 678 00:26:27,640 --> 00:26:29,349 testing to a list of sites created by 679 00:26:29,350 --> 00:26:30,350 country experts. 680 00:26:31,860 --> 00:26:33,929 Most of the sites that we found to be 681 00:26:33,930 --> 00:26:36,029 blocked include news 682 00:26:36,030 --> 00:26:38,159 outlets and blogs, as well 683 00:26:38,160 --> 00:26:40,590 as pro-democracy and human rights sites, 684 00:26:41,760 --> 00:26:43,919 many of them seem to have one main thing 685 00:26:43,920 --> 00:26:44,969 in common. 686 00:26:44,970 --> 00:26:47,759 Directly or indirectly, 687 00:26:47,760 --> 00:26:49,979 they express criticism towards the Cuban 688 00:26:49,980 --> 00:26:51,269 government. 689 00:26:51,270 --> 00:26:53,009 Other circumvention tools, like the Tor 690 00:26:53,010 --> 00:26:54,539 anonymity network were found to be 691 00:26:54,540 --> 00:26:55,760 accessible across Cuba. 692 00:26:57,110 --> 00:26:58,999 Our software not only allowed us to 693 00:26:59,000 --> 00:27:01,279 confirm whether sites were blocked, 694 00:27:01,280 --> 00:27:02,869 but it also allowed us to collect network 695 00:27:02,870 --> 00:27:04,939 measurement data that clearly shows how 696 00:27:04,940 --> 00:27:06,649 censorship is implemented. 697 00:27:06,650 --> 00:27:08,569 ETECSA appears to be blocking sites 698 00:27:08,570 --> 00:27:10,009 through the use of technologies that 699 00:27:10,010 --> 00:27:12,499 resets connections or service block pages 700 00:27:12,500 --> 00:27:14,059 instead of the intended Web page. 701 00:27:16,200 --> 00:27:18,449 In many countries around the world, when 702 00:27:18,450 --> 00:27:20,879 a website is blocked, you know it 703 00:27:20,880 --> 00:27:22,980 because it often looks something like. 704 00:27:24,270 --> 00:27:26,250 This or this, 705 00:27:27,810 --> 00:27:29,429 black pages usually include some 706 00:27:29,430 --> 00:27:31,619 information not notifying you what, 707 00:27:31,620 --> 00:27:33,179 that the site is blocked and sometimes 708 00:27:33,180 --> 00:27:34,649 even include a reference to the legal 709 00:27:34,650 --> 00:27:36,719 justification, which means that 710 00:27:36,720 --> 00:27:38,609 it may not always be clear that the site 711 00:27:38,610 --> 00:27:40,409 is intentionally being blocked. 712 00:27:40,410 --> 00:27:42,599 Users might assume that 713 00:27:42,600 --> 00:27:44,399 it's a connectivity issue rather than 714 00:27:44,400 --> 00:27:46,769 government commission censorship. 715 00:27:46,770 --> 00:27:48,809 But it isn't the only ISP in the world to 716 00:27:48,810 --> 00:27:51,059 serve Black Bloc pages. 717 00:27:51,060 --> 00:27:53,659 You see this in many countries as well. 718 00:27:53,660 --> 00:27:55,369 Interestingly enough, we only found the 719 00:27:55,370 --> 00:27:57,290 HTP version of the sites to be blocked. 720 00:27:59,080 --> 00:28:01,389 And the GPS is unblocked, meaning 721 00:28:01,390 --> 00:28:03,759 that people could circumvented 722 00:28:03,760 --> 00:28:06,099 using a GPS, however, many sites 723 00:28:06,100 --> 00:28:07,569 that we found to be blocked do not 724 00:28:07,570 --> 00:28:08,570 support a 725 00:28:10,060 --> 00:28:12,159 Chinese vendor. Huawei appears to be 726 00:28:12,160 --> 00:28:13,749 supporting Cuba's Internet 727 00:28:13,750 --> 00:28:15,609 infrastructure, and you don't even need 728 00:28:15,610 --> 00:28:17,619 to run any tests to find this out. 729 00:28:17,620 --> 00:28:19,359 The server header of the Black Pages, for 730 00:28:19,360 --> 00:28:21,639 example, points to Huawei equipment. 731 00:28:21,640 --> 00:28:24,669 Even a Texas captive portal itself 732 00:28:24,670 --> 00:28:26,709 seems to be written by Chinese developers 733 00:28:26,710 --> 00:28:28,539 since they left over some comments in the 734 00:28:28,540 --> 00:28:30,430 source code of the captive portal page. 735 00:28:31,920 --> 00:28:34,119 Overall, Internet censorship in Cuba does 736 00:28:34,120 --> 00:28:35,379 not appear to be particularly 737 00:28:35,380 --> 00:28:37,479 sophisticated given the high costs 738 00:28:37,480 --> 00:28:38,979 of the Internet, rendering it 739 00:28:38,980 --> 00:28:40,479 inaccessible to most Cubans. 740 00:28:40,480 --> 00:28:42,009 Perhaps the government doesn't even need 741 00:28:42,010 --> 00:28:44,169 to invest in sophisticated Internet 742 00:28:44,170 --> 00:28:45,399 censorship. 743 00:28:45,400 --> 00:28:46,899 But Cuba's Internet landscape is 744 00:28:46,900 --> 00:28:49,239 changing, and so might Internet 745 00:28:49,240 --> 00:28:50,240 censorship. 746 00:28:53,540 --> 00:28:55,279 That's why we think it's important to 747 00:28:55,280 --> 00:28:57,769 monitor networks continuously, users 748 00:28:57,770 --> 00:28:59,869 in Cuba and around the world can 749 00:28:59,870 --> 00:29:01,999 run any probe to shed light 750 00:29:02,000 --> 00:29:03,349 on information controls. 751 00:29:04,710 --> 00:29:05,710 Thank you. 752 00:29:19,670 --> 00:29:21,859 A, I think I finished right 753 00:29:21,860 --> 00:29:24,199 on the dot, so I suspect 754 00:29:24,200 --> 00:29:26,479 that thank 755 00:29:26,480 --> 00:29:28,429 you very much. So everybody in the 756 00:29:28,430 --> 00:29:30,499 audience who now got curious about 757 00:29:30,500 --> 00:29:32,419 money, maybe you want to consider if you 758 00:29:32,420 --> 00:29:34,459 want to contribute by running an uni 759 00:29:34,460 --> 00:29:36,349 probe in your own country or when you're 760 00:29:36,350 --> 00:29:39,019 on holidays and some interesting places, 761 00:29:39,020 --> 00:29:40,369 just think about whether you would like 762 00:29:40,370 --> 00:29:42,679 to contribute and help out. 763 00:29:42,680 --> 00:29:44,899 And we have an assembly on in 764 00:29:44,900 --> 00:29:46,879 the CCL and the second floor, that 765 00:29:46,880 --> 00:29:48,919 universe. If you have more questions 766 00:29:48,920 --> 00:29:50,659 about money, you should come and join us 767 00:29:50,660 --> 00:29:52,729 upstairs and we will be 768 00:29:52,730 --> 00:29:54,829 happy to answer any questions you may 769 00:29:54,830 --> 00:29:55,769 have. Exactly. 770 00:29:55,770 --> 00:29:57,289 Perfect. Check them out and their only 771 00:29:57,290 --> 00:29:59,059 assembly, please give another big round 772 00:29:59,060 --> 00:30:00,209 of applause to approval. 773 00:30:00,210 --> 00:30:01,210 Thank you very much.