1 00:00:09,960 --> 00:00:14,400 Speaker 1: In the nineteen nineties, we had this idea that the 2 00:00:14,440 --> 00:00:17,079 Speaker 1: Internet was just this open space and we could try 3 00:00:17,480 --> 00:00:21,520 Speaker 1: all sorts of nonsense there. Piero Midjar wants to sell 4 00:00:21,520 --> 00:00:24,959 Speaker 1: his pez collection, and you end up with eBay, and 5 00:00:25,000 --> 00:00:27,120 Speaker 1: then people realize that people are ripping each other off 6 00:00:27,120 --> 00:00:29,160 Speaker 1: on eBay, and so you end up with an escro service, 7 00:00:29,200 --> 00:00:32,680 Speaker 1: and that turns into PayPal, and like, all these things happen, 8 00:00:33,120 --> 00:00:36,600 Speaker 1: and they happen really really fast. I want us to 9 00:00:36,640 --> 00:00:40,479 Speaker 1: get back to that moment of thinking things are possible. 10 00:00:41,360 --> 00:00:45,000 Speaker 2: Ethan Zuckerman is a professor at the University of Massachusetts Amherst. 11 00:00:45,240 --> 00:00:47,919 Speaker 2: He's also something of an Internet activist, and. 12 00:00:47,880 --> 00:00:50,479 Speaker 1: I work on a whole lot of projects trying to 13 00:00:50,479 --> 00:00:53,240 Speaker 1: make the Internet a better place, which I still believe 14 00:00:53,320 --> 00:00:55,680 Speaker 1: can happen despite all the evidence of the contrary. 15 00:00:56,000 --> 00:00:57,720 Speaker 2: You know what, you may be one of the last 16 00:00:57,760 --> 00:00:59,360 Speaker 2: soldiers standing on that battlefield. 17 00:00:59,400 --> 00:01:02,600 Speaker 1: I know, man, I know, I know it's a lonely 18 00:01:02,640 --> 00:01:04,080 Speaker 1: place to be. But you know, the great thing about 19 00:01:04,080 --> 00:01:06,560 Speaker 1: being a professor is that I can try to win 20 00:01:06,680 --> 00:01:10,520 Speaker 1: over students to the cause, and every so often, you know, 21 00:01:10,560 --> 00:01:13,120 Speaker 1: we get a win. Every so often. People are excited 22 00:01:13,120 --> 00:01:18,440 Speaker 1: about this idea that things actually can change, and I 23 00:01:18,520 --> 00:01:19,880 Speaker 1: think a lot of it just hass to do with 24 00:01:20,040 --> 00:01:21,080 Speaker 1: reframing our thinking. 25 00:01:21,600 --> 00:01:25,039 Speaker 2: Ethan is one of the most unusual professors I've ever 26 00:01:25,080 --> 00:01:28,040 Speaker 2: spoken to. So, first off, if you're of a certain age, 27 00:01:28,120 --> 00:01:31,080 Speaker 2: I guarantee you that you've seen his work. It might 28 00:01:31,120 --> 00:01:33,160 Speaker 2: have made you mad, It might have made you money, 29 00:01:33,319 --> 00:01:35,720 Speaker 2: it might have done both at the same time. The second, 30 00:01:36,000 --> 00:01:40,040 Speaker 2: Ethan really really likes Facebook, so much so that he 31 00:01:40,200 --> 00:01:43,080 Speaker 2: sued Facebook, and you know what, after listening to what 32 00:01:43,120 --> 00:01:45,280 Speaker 2: he has to say, you might want him to win 33 00:01:50,200 --> 00:01:57,680 Speaker 2: from Kaleidoscope and iHeart podcasts. This is kill Switch. I'm 34 00:01:57,720 --> 00:02:00,000 Speaker 2: Dexter Thomas. 35 00:02:04,800 --> 00:02:42,160 Speaker 3: I'm sorry, I'm sorry. Good bye. 36 00:02:46,720 --> 00:02:50,040 Speaker 2: Ethan Zuckerman wants to fix the Internet. But before we 37 00:02:50,040 --> 00:02:51,720 Speaker 2: could talk about how he would do that, we need 38 00:02:51,720 --> 00:02:56,000 Speaker 2: to understand how we got here. We've talked about this 39 00:02:56,080 --> 00:02:58,560 Speaker 2: here before on the show, but early on the Internet 40 00:02:58,639 --> 00:03:02,560 Speaker 2: was a much more optimistic place until money basically ruined it. 41 00:03:02,880 --> 00:03:05,320 Speaker 2: If you haven't caught our episode with Tim, we'll definitely 42 00:03:05,400 --> 00:03:08,760 Speaker 2: check that one out too. So Ethan started his career 43 00:03:08,840 --> 00:03:10,920 Speaker 2: during those early dot com days. He was one of 44 00:03:10,919 --> 00:03:14,520 Speaker 2: the first employees at a company called Tripod, and Tripod 45 00:03:14,600 --> 00:03:16,120 Speaker 2: was most well known for being a place where you 46 00:03:16,120 --> 00:03:18,760 Speaker 2: could build your own homepage, which back then was a 47 00:03:18,800 --> 00:03:22,040 Speaker 2: really exciting thing. There were all kinds of communities popping 48 00:03:22,120 --> 00:03:25,560 Speaker 2: up online and Tripod was helping to foster that. But 49 00:03:25,600 --> 00:03:28,200 Speaker 2: in order to do that, they needed money, and so 50 00:03:28,440 --> 00:03:33,120 Speaker 2: they started looking at advertising. I was a GeoCities person. Yep, 51 00:03:33,200 --> 00:03:35,600 Speaker 2: that's what I was on. We're talking pre my Space here, 52 00:03:35,640 --> 00:03:36,520 Speaker 2: but I remember. 53 00:03:36,240 --> 00:03:38,760 Speaker 1: We're talking pre my Space and that was our leading 54 00:03:38,800 --> 00:03:41,160 Speaker 1: competitor at Tripods. So I'm not going to hold that 55 00:03:41,200 --> 00:03:44,480 Speaker 1: against you, Dexter, but you know, that is a bit 56 00:03:44,520 --> 00:03:46,200 Speaker 1: of a you know, you did just a little bit 57 00:03:46,200 --> 00:03:48,520 Speaker 1: of a Yankees Mets thing there. You gotta be careful about. 58 00:03:48,560 --> 00:03:51,600 Speaker 2: God, Hey, listen, I'm just trying to be honest with you. 59 00:03:51,640 --> 00:03:53,640 Speaker 1: I'm just fair enough. That's where I went. 60 00:03:53,920 --> 00:03:57,520 Speaker 2: But you know, I mean, as I was entering this 61 00:03:57,680 --> 00:04:00,520 Speaker 2: phase of Wow, there's all these other people made things, 62 00:04:00,520 --> 00:04:04,000 Speaker 2: and I would like to make things too. Very early 63 00:04:04,040 --> 00:04:06,600 Speaker 2: in that stage, I realized that that was always going 64 00:04:06,640 --> 00:04:10,160 Speaker 2: to mean there's gonna be ads next to whatever I did, 65 00:04:10,600 --> 00:04:13,680 Speaker 2: or on top of whatever I did, or inside of 66 00:04:13,720 --> 00:04:17,920 Speaker 2: whatever I did. And that started to feel very normal. 67 00:04:18,360 --> 00:04:22,479 Speaker 1: Yeah, So like roll back to GeoCities and I'm going 68 00:04:22,520 --> 00:04:23,880 Speaker 1: to tell you what's going on a tripod. 69 00:04:23,960 --> 00:04:24,120 Speaker 3: Right. 70 00:04:24,200 --> 00:04:27,320 Speaker 1: So, so our biggest rival geo cities is West Coast. 71 00:04:27,360 --> 00:04:30,280 Speaker 1: We're East Coast. So there's beef. We know each other, 72 00:04:30,480 --> 00:04:33,000 Speaker 1: we're you know, we're in touch with one another. But 73 00:04:33,160 --> 00:04:36,800 Speaker 1: we're all trying the same ideas. We try a subscription service. 74 00:04:36,839 --> 00:04:38,800 Speaker 1: We've got it. You know, you can sign up, you 75 00:04:38,800 --> 00:04:42,120 Speaker 1: can get extra space, you can get premium customer service. 76 00:04:42,480 --> 00:04:46,320 Speaker 1: Nobody wants that. We try merch We make the tea shirts, 77 00:04:46,360 --> 00:04:48,520 Speaker 1: we make the hats. We're psyched that people want to 78 00:04:48,520 --> 00:04:52,080 Speaker 1: identify with Tripod. We think that's a way that we're 79 00:04:52,080 --> 00:04:55,880 Speaker 1: going to go. We go with sponsorship. We actually have 80 00:04:56,000 --> 00:04:59,080 Speaker 1: sponsors who are like these are probably high school and 81 00:04:59,080 --> 00:05:01,760 Speaker 1: college students. You know what high school and college students 82 00:05:01,839 --> 00:05:05,200 Speaker 1: really love is mutual funds. Calvert Mutual Funds comes on 83 00:05:05,400 --> 00:05:08,240 Speaker 1: and says, we want to sponsor people to go and 84 00:05:08,240 --> 00:05:11,120 Speaker 1: get mutual funds. The only thing that works for us 85 00:05:11,200 --> 00:05:17,479 Speaker 1: dexter is advertising, and the whole industry moves there at 86 00:05:17,520 --> 00:05:21,039 Speaker 1: the same time and at a certain point, if you're 87 00:05:21,120 --> 00:05:24,480 Speaker 1: trying something new, if you're trying a different business model, 88 00:05:25,320 --> 00:05:29,479 Speaker 1: your pitch to an investor gets very tricky. What the 89 00:05:29,560 --> 00:05:32,400 Speaker 1: investors want to hear is how many page views are 90 00:05:32,400 --> 00:05:34,360 Speaker 1: you getting a month and how fast is it growing? 91 00:05:35,080 --> 00:05:39,720 Speaker 1: And because we have eighteen million people like you who 92 00:05:39,839 --> 00:05:42,640 Speaker 1: put home pages on the site, we are getting the 93 00:05:42,760 --> 00:05:46,640 Speaker 1: eighth most page views on the Internet and it's growing 94 00:05:46,760 --> 00:05:50,760 Speaker 1: something like ten percent a month. And companies like Yahoo 95 00:05:51,000 --> 00:05:53,320 Speaker 1: are coming and knocking on our door and saying we 96 00:05:53,320 --> 00:05:57,520 Speaker 1: would like to do some business. We are like, you know, 97 00:05:57,880 --> 00:06:01,360 Speaker 1: kids in our twenties in the top floor of a 98 00:06:01,760 --> 00:06:07,600 Speaker 1: plastic factory in Williamstown, Massachusetts, and like Jerry Yang from 99 00:06:07,680 --> 00:06:10,320 Speaker 1: Yahoo is calling us, and we're like, yeah, I guess 100 00:06:10,400 --> 00:06:11,520 Speaker 1: advertising is the way to go. 101 00:06:12,320 --> 00:06:14,839 Speaker 2: So advertising is the way to go might not be 102 00:06:14,920 --> 00:06:19,479 Speaker 2: the most idealistic phrase ever, but again, advertising made Tripod work, 103 00:06:19,760 --> 00:06:22,520 Speaker 2: and it also made some pretty interesting things possible. 104 00:06:23,040 --> 00:06:25,440 Speaker 1: The other thing it's an amazing model for is for 105 00:06:25,640 --> 00:06:31,400 Speaker 1: people who are unbanked, who are not part of the 106 00:06:31,440 --> 00:06:35,680 Speaker 1: mainstream financial system, who are in countries where the amounts 107 00:06:35,680 --> 00:06:39,600 Speaker 1: of money would be very challenging. One thing you probably 108 00:06:39,600 --> 00:06:41,320 Speaker 1: did not know when you were fooling around with that 109 00:06:41,360 --> 00:06:46,000 Speaker 1: Tripod homepage is that our number two country for users 110 00:06:46,080 --> 00:06:48,760 Speaker 1: was Malaysia. And what was going on in Malaysia at 111 00:06:48,760 --> 00:06:53,240 Speaker 1: that point was most of the political opposition was using 112 00:06:53,320 --> 00:06:58,159 Speaker 1: Tripod to organize and it became the voice of the 113 00:06:58,200 --> 00:07:02,960 Speaker 1: Malaysian REFORMASSEI movement, And those folks might have been able 114 00:07:03,279 --> 00:07:08,600 Speaker 1: to afford official websites somewhere, but they weren't really set 115 00:07:08,680 --> 00:07:12,600 Speaker 1: up for transacting with Malaysian users. They probably didn't have 116 00:07:12,640 --> 00:07:15,800 Speaker 1: credit cards, they didn't have the payment mechanisms. So advertising 117 00:07:15,880 --> 00:07:19,280 Speaker 1: is a great way to help out people who are 118 00:07:19,400 --> 00:07:22,680 Speaker 1: coming from different parts of the world, who have different resources. 119 00:07:23,280 --> 00:07:27,120 Speaker 2: It wasn't just Tripod. Advertising became the default model for 120 00:07:27,160 --> 00:07:30,840 Speaker 2: supporting publishing on the Internet, which shouldn't be surprising. They 121 00:07:30,840 --> 00:07:33,760 Speaker 2: were just adapting what print magazines and newspapers were doing. 122 00:07:34,360 --> 00:07:38,000 Speaker 1: Ads at this point are being sold by what's called CPM, 123 00:07:38,440 --> 00:07:42,440 Speaker 1: and CPM is cost per mill which you know, confusingly enough, 124 00:07:42,480 --> 00:07:45,880 Speaker 1: is cost per thousand, right. It's a term coming out 125 00:07:45,920 --> 00:07:50,360 Speaker 1: of the newspaper and magazine industry, And so if you 126 00:07:50,440 --> 00:07:54,640 Speaker 1: want to buy a page in Wired, for instance, you're 127 00:07:54,720 --> 00:07:58,640 Speaker 1: basically being told by Wired, well, here's our circulation. You know, 128 00:07:58,760 --> 00:08:03,320 Speaker 1: we're reaching fifty thousand people multiply that out by the 129 00:08:03,480 --> 00:08:06,840 Speaker 1: cost per thousand that's what you're gonna pay for a page. 130 00:08:06,840 --> 00:08:09,840 Speaker 1: And Wired we come in nineteen ninety five and go, 131 00:08:10,360 --> 00:08:12,480 Speaker 1: we are so much better than that page. And Wired 132 00:08:12,920 --> 00:08:14,680 Speaker 1: you can click on our ad and it will take 133 00:08:14,680 --> 00:08:17,800 Speaker 1: you to your website and people might spend hours with you, 134 00:08:17,960 --> 00:08:20,600 Speaker 1: not just that moment on the page, and you can 135 00:08:20,640 --> 00:08:23,680 Speaker 1: take their name and address. So we are worth so 136 00:08:23,920 --> 00:08:27,520 Speaker 1: much more than that magazine ad. So we start out 137 00:08:27,720 --> 00:08:29,960 Speaker 1: pricing the way that magazine ads are. We're is charging 138 00:08:30,000 --> 00:08:34,720 Speaker 1: like fifty bucks for one thousand impressions. And then people 139 00:08:34,840 --> 00:08:37,000 Speaker 1: realize that there's ads all. 140 00:08:36,880 --> 00:08:37,480 Speaker 2: Over the web. 141 00:08:38,160 --> 00:08:41,120 Speaker 1: You can go wherever you want and get ad inventory, 142 00:08:41,600 --> 00:08:44,160 Speaker 1: and so suddenly it drops to like a dollar per 143 00:08:44,240 --> 00:08:48,880 Speaker 1: thousand impressions. And then Overture, which is a company that's 144 00:08:48,880 --> 00:08:52,400 Speaker 1: affiliated with Yahoo, comes in and says, you know, the 145 00:08:52,440 --> 00:08:55,560 Speaker 1: ads that you don't click on, those shouldn't count, and 146 00:08:55,640 --> 00:08:58,600 Speaker 1: so then it moves to cost per click, and that's 147 00:08:58,600 --> 00:09:03,920 Speaker 1: an entirely different space. In the course of five years, 148 00:09:04,480 --> 00:09:07,360 Speaker 1: we go from ads being fifty two one hundred dollars 149 00:09:07,360 --> 00:09:12,160 Speaker 1: for one thousand impressions to being twenty five cents for 150 00:09:12,200 --> 00:09:17,440 Speaker 1: a thousand impressions to being impressions don't matter. We will 151 00:09:17,480 --> 00:09:20,600 Speaker 1: pay anywhere from a penny to ten bucks per collect 152 00:09:20,600 --> 00:09:21,720 Speaker 1: depending on what the ad is. 153 00:09:22,800 --> 00:09:25,600 Speaker 2: And you know where this goes. Ads not only became 154 00:09:25,640 --> 00:09:28,560 Speaker 2: an escapable on the Internet, but as clicks become the 155 00:09:28,600 --> 00:09:31,679 Speaker 2: metric of success, you have to start listening to what 156 00:09:31,720 --> 00:09:34,480 Speaker 2: the advertisers want. And so here's the thing about Ethan. 157 00:09:34,840 --> 00:09:37,640 Speaker 2: He wasn't just a pioneer in the early dot com boom. 158 00:09:37,679 --> 00:09:40,360 Speaker 2: He helped facilitate it by figuring out a way to 159 00:09:40,440 --> 00:09:43,440 Speaker 2: keep those advertisers shilling out that money for our websites. 160 00:09:43,880 --> 00:09:47,240 Speaker 2: He essentially invented was now known as the pop up ad. 161 00:09:47,679 --> 00:09:50,760 Speaker 2: To be clear, he regrets that he did this the 162 00:09:50,760 --> 00:09:53,120 Speaker 2: innovation of pop up ads, and we should call it 163 00:09:53,160 --> 00:09:53,680 Speaker 2: an innovation. 164 00:09:54,160 --> 00:09:56,040 Speaker 1: I think I'm not going to let you get away 165 00:09:56,080 --> 00:10:00,000 Speaker 1: with innovation. First of all, come up ads no, no, no, no, sorry, 166 00:10:00,760 --> 00:10:04,400 Speaker 1: I'm sorry. Pop up ads are, to use a technical term, 167 00:10:04,600 --> 00:10:10,079 Speaker 1: a Cluegi hack. It is an inelegant solution to a problem, 168 00:10:10,480 --> 00:10:15,199 Speaker 1: and unfortunately, like a lot of cluegi hacks, it still persists, 169 00:10:15,240 --> 00:10:17,240 Speaker 1: although thank god, not as much as it used to. 170 00:10:17,600 --> 00:10:20,079 Speaker 1: So here's the story on the pop up bed. Let's 171 00:10:20,080 --> 00:10:24,559 Speaker 1: go back to nineteen ninety six. We've got an increasing 172 00:10:24,640 --> 00:10:28,520 Speaker 1: number of user homepages. They have become the center of 173 00:10:28,559 --> 00:10:31,040 Speaker 1: our business. Tripod thought it was going to be like 174 00:10:31,120 --> 00:10:34,160 Speaker 1: a magazine and educational site for young people. My friend 175 00:10:34,200 --> 00:10:37,720 Speaker 1: Jeff Vanderklute, without my permission, I should add, puts up 176 00:10:37,720 --> 00:10:41,559 Speaker 1: a homepage builder. We discover a couple months later, when 177 00:10:41,559 --> 00:10:44,400 Speaker 1: our bandwidth bill has gone through the ceiling, that it 178 00:10:44,440 --> 00:10:46,840 Speaker 1: is the most popular thing on our site. So it's 179 00:10:46,880 --> 00:10:49,560 Speaker 1: completely accidental. We get into the business completely by accident. 180 00:10:50,160 --> 00:10:54,480 Speaker 1: But by nineteen ninety six, my boss Bo Peebet understands 181 00:10:54,520 --> 00:10:56,480 Speaker 1: that this is the tail that's wagging the dog. He 182 00:10:56,600 --> 00:10:58,880 Speaker 1: understands that this is where our growth is coming from. 183 00:10:59,520 --> 00:11:01,959 Speaker 1: And he's trying to sell ad inventory on this. And 184 00:11:02,000 --> 00:11:06,800 Speaker 1: we've got ad execs going out and they are showing 185 00:11:07,400 --> 00:11:11,600 Speaker 1: random homepages with ads on them, and often those ads 186 00:11:11,720 --> 00:11:15,280 Speaker 1: are showing up on content that the advertisers do not 187 00:11:15,400 --> 00:11:20,040 Speaker 1: want to be associated with. Sometimes it's extremism, sometimes it's pornography. 188 00:11:20,920 --> 00:11:23,640 Speaker 1: It is the web, right and unfortunately, you know, the 189 00:11:23,679 --> 00:11:26,920 Speaker 1: Internet has been an if he placed going back into 190 00:11:26,920 --> 00:11:30,720 Speaker 1: the nineteen seventies, so Bo comes to me and says, look, 191 00:11:31,080 --> 00:11:34,079 Speaker 1: we got to run ads against the homepages. But as 192 00:11:34,200 --> 00:11:37,560 Speaker 1: long as the ad is on the homepage, the advertiser 193 00:11:37,600 --> 00:11:42,000 Speaker 1: feels like they're endorsing them. What can you do? And 194 00:11:42,160 --> 00:11:46,840 Speaker 1: I invent the nav bar. The navbar is a vertical panel. 195 00:11:47,640 --> 00:11:50,160 Speaker 1: It has a two hundred by two hundred add on it, 196 00:11:50,160 --> 00:11:54,040 Speaker 1: It has Tripod branding, it has some navigation links, and 197 00:11:54,080 --> 00:11:57,840 Speaker 1: so it's like, hey, here's Dexter's homepage, but also here's 198 00:11:57,840 --> 00:12:01,880 Speaker 1: some Tripod branding, some navigation. Oh and by the way, 199 00:12:01,960 --> 00:12:05,800 Speaker 1: an ad. And they're separate windows. So it's very clear 200 00:12:05,840 --> 00:12:09,079 Speaker 1: that the ad is against Tripod and your homepage is 201 00:12:09,120 --> 00:12:09,640 Speaker 1: your space. 202 00:12:10,320 --> 00:12:13,040 Speaker 2: So the pop up ad was created to solve a problem. 203 00:12:13,280 --> 00:12:16,560 Speaker 2: Advertisers were scared that their ads would show up somewhere unsavory, 204 00:12:16,960 --> 00:12:19,160 Speaker 2: and the pop up helped to alleviate that fear by 205 00:12:19,200 --> 00:12:22,520 Speaker 2: creating that separation so that they would keep that money flowing. 206 00:12:23,040 --> 00:12:25,400 Speaker 1: And I'm like, this is great, I've worked it out. 207 00:12:25,559 --> 00:12:28,920 Speaker 1: It's a solution. We put up the little piece of JavaScript, 208 00:12:28,960 --> 00:12:31,080 Speaker 1: we insert it into every one of the homepages that's 209 00:12:31,080 --> 00:12:33,480 Speaker 1: out there. We got a little pushback. People are sort 210 00:12:33,480 --> 00:12:35,480 Speaker 1: of like, you know, what's going on? Why is there 211 00:12:35,520 --> 00:12:38,160 Speaker 1: this extra weird window here, But we don't get a 212 00:12:38,200 --> 00:12:42,760 Speaker 1: ton of pushback. Your guys at GeoCities take my code, 213 00:12:43,400 --> 00:12:46,600 Speaker 1: they turn it from vertical into horizontal. They take my 214 00:12:46,920 --> 00:12:50,040 Speaker 1: polite little two hundred by two hundred AD and they 215 00:12:50,080 --> 00:12:53,120 Speaker 1: turn it into you a full withd like seven hundred 216 00:12:53,120 --> 00:12:56,640 Speaker 1: pixel AD and there's no branding or anything. It's just 217 00:12:56,720 --> 00:13:01,560 Speaker 1: an AD. So I do the first pop up, and 218 00:13:01,640 --> 00:13:04,040 Speaker 1: it is a pop up AD. So I'm absolutely owning 219 00:13:04,120 --> 00:13:06,160 Speaker 1: up to it. But it was trying to be maybe 220 00:13:06,200 --> 00:13:09,080 Speaker 1: a bit more than that. The essence of it gets 221 00:13:09,120 --> 00:13:12,920 Speaker 1: stripped down to a pop up AD, and then it's 222 00:13:12,960 --> 00:13:15,240 Speaker 1: off to the races, right like then people are doing 223 00:13:15,280 --> 00:13:18,160 Speaker 1: these pop unders. They're doing pop ups that can't be minimized, 224 00:13:18,160 --> 00:13:22,360 Speaker 1: so on and so forth, and you have really the 225 00:13:22,559 --> 00:13:26,719 Speaker 1: scourge of the nineteen nineties and early two thousands internet. 226 00:13:27,160 --> 00:13:29,400 Speaker 1: And let me just say just to be clear on this. 227 00:13:30,160 --> 00:13:32,720 Speaker 1: When we were talking before this interview, I told you 228 00:13:32,880 --> 00:13:35,120 Speaker 1: I did not want to do an interview on pop 229 00:13:35,200 --> 00:13:37,559 Speaker 1: up ads. And part of it is that it's a 230 00:13:37,600 --> 00:13:40,840 Speaker 1: little too easy. It's a little like you know, look, 231 00:13:40,840 --> 00:13:43,679 Speaker 1: it's the Robert Oppenheimer of the Internet who hates his creation. 232 00:13:44,600 --> 00:13:48,400 Speaker 1: Let me be honest with you, Dexter, I get hate mail. 233 00:13:48,720 --> 00:13:52,240 Speaker 1: Every time someone writes a story. I get death threats. 234 00:13:52,920 --> 00:13:57,439 Speaker 1: I have a relationship with my local police department in 235 00:13:57,480 --> 00:14:01,680 Speaker 1: this small town in western Massachusetts where they know not 236 00:14:01,800 --> 00:14:04,840 Speaker 1: to call a swap team to my residence because almost 237 00:14:04,880 --> 00:14:07,320 Speaker 1: every time I do an interview on pop ups that 238 00:14:07,360 --> 00:14:11,080 Speaker 1: gets some exposure, we get an arm response call. So 239 00:14:11,920 --> 00:14:15,520 Speaker 1: it's a thing, and it's interesting. It's definitely the sort 240 00:14:15,520 --> 00:14:17,839 Speaker 1: of thing that I did not anticipate, either when building 241 00:14:17,880 --> 00:14:22,280 Speaker 1: the pop up or twenty years later, owning up to 242 00:14:22,320 --> 00:14:25,720 Speaker 1: building the pop up and talking about it as a problem. 243 00:14:26,560 --> 00:14:29,600 Speaker 2: Seeing all the anger that gets directed in Ethan is. 244 00:14:29,720 --> 00:14:33,120 Speaker 2: Aside from being really terrifying, it's also kind of confusing 245 00:14:33,160 --> 00:14:37,480 Speaker 2: because really you're mad about pop ups in twenty twenty six, like, 246 00:14:37,520 --> 00:14:39,480 Speaker 2: are you still dialing up from your mom's dell in 247 00:14:39,520 --> 00:14:40,200 Speaker 2: the living room? 248 00:14:40,720 --> 00:14:40,920 Speaker 1: Yell? 249 00:14:40,920 --> 00:14:43,160 Speaker 2: Are mad at the wrong dude? Because first off, pop 250 00:14:43,240 --> 00:14:45,200 Speaker 2: ups didn't even last that long. We're going to come 251 00:14:45,200 --> 00:14:48,320 Speaker 2: back to that later, but the industry has long since 252 00:14:48,360 --> 00:14:51,040 Speaker 2: moved on to something that has zero to do with 253 00:14:51,080 --> 00:14:54,400 Speaker 2: what anyone at tripod dot com could ever even thought 254 00:14:54,400 --> 00:14:57,600 Speaker 2: of back then. Traditional banner ads, which again, this is 255 00:14:57,600 --> 00:15:00,800 Speaker 2: the thing that Ethan was trying to replace. Those came back, 256 00:15:01,160 --> 00:15:04,600 Speaker 2: and they came back stronger. Remember, the original problem was 257 00:15:04,640 --> 00:15:07,480 Speaker 2: that banner ads might be irrelevant to the website that 258 00:15:07,560 --> 00:15:10,240 Speaker 2: they were on, and thus irrelevant to you, the user. 259 00:15:10,840 --> 00:15:13,080 Speaker 2: But the industry figured out how to make those ads 260 00:15:13,160 --> 00:15:17,360 Speaker 2: much more relevant to you by watching everything you do 261 00:15:17,480 --> 00:15:20,680 Speaker 2: on the Internet. And this is where Ethan started thinking 262 00:15:20,880 --> 00:15:21,680 Speaker 2: a little differently. 263 00:15:22,640 --> 00:15:26,960 Speaker 1: The problem with advertising, as I see it, is that 264 00:15:27,000 --> 00:15:29,120 Speaker 1: the people who have been trying to make money out 265 00:15:29,120 --> 00:15:34,840 Speaker 1: of advertising have tried to add every possible piece of 266 00:15:34,880 --> 00:15:38,960 Speaker 1: information to the equation. They want to know where you are, 267 00:15:39,440 --> 00:15:41,440 Speaker 1: They want to know where else on the Internet that 268 00:15:41,480 --> 00:15:44,160 Speaker 1: you've been, They want to know what you've done, what 269 00:15:44,200 --> 00:15:48,000 Speaker 1: you've searched for previously. And so they turned this Internet, 270 00:15:48,040 --> 00:15:50,960 Speaker 1: this super exciting space that we're building in the nineteen nineties, 271 00:15:51,480 --> 00:15:56,120 Speaker 1: into a surveillance machine. My friend Chan Regenda Nikolucci, and 272 00:15:56,160 --> 00:15:59,600 Speaker 1: I started looking at this and saying, what if you 273 00:15:59,680 --> 00:16:05,200 Speaker 1: could get eighty percent of what you wanted with so 274 00:16:05,400 --> 00:16:06,760 Speaker 1: much less of the surveillance. 275 00:16:08,520 --> 00:16:11,200 Speaker 2: Suddenly Ethan had a new mission to keep the lights 276 00:16:11,240 --> 00:16:14,120 Speaker 2: on the Internet without having to sell out your audience. 277 00:16:15,240 --> 00:16:27,040 Speaker 2: That's after the break. So years later, Ethan sees how 278 00:16:27,080 --> 00:16:30,600 Speaker 2: the Internet has grown into a surveillance machine, partially in 279 00:16:30,720 --> 00:16:34,120 Speaker 2: service of the advertising that still drives the Internet, and 280 00:16:34,160 --> 00:16:36,880 Speaker 2: he started to think, could we do anything about this? 281 00:16:37,520 --> 00:16:40,480 Speaker 1: My attitude in some ways is, well, can't we just 282 00:16:40,880 --> 00:16:41,720 Speaker 1: end advertising? 283 00:16:41,800 --> 00:16:42,080 Speaker 3: Right? Like? 284 00:16:42,120 --> 00:16:46,800 Speaker 1: Advertisings a terrible model. We're paying with our attention. But 285 00:16:46,880 --> 00:16:49,560 Speaker 1: here's the thing about ads. Ads turn out to be 286 00:16:50,680 --> 00:16:56,000 Speaker 1: a fantastic model for a product or service that you 287 00:16:56,120 --> 00:16:58,400 Speaker 1: don't know that you're willing to pay for yet. So 288 00:16:58,680 --> 00:17:04,040 Speaker 1: imagine that someone is launching something new. Right, Someone in 289 00:17:04,080 --> 00:17:07,560 Speaker 1: Western Massachusetts where I live, says, I don't like the 290 00:17:07,600 --> 00:17:09,639 Speaker 1: way local media is working. I'm going to create a 291 00:17:09,640 --> 00:17:13,359 Speaker 1: new news product. It's going to be video first, and 292 00:17:13,400 --> 00:17:16,040 Speaker 1: it's going to be put together by citizens. And I'm like, well, 293 00:17:16,040 --> 00:17:18,720 Speaker 1: that might be interesting, but I'm not ready to put 294 00:17:18,800 --> 00:17:21,560 Speaker 1: up ten bucks a month for it. Let me see 295 00:17:21,560 --> 00:17:24,360 Speaker 1: how it's going first, and under the let me see 296 00:17:24,359 --> 00:17:27,040 Speaker 1: how it's going first. Advertising is a great model. 297 00:17:27,720 --> 00:17:30,320 Speaker 2: Okay, It's probably not realistic to think that the Internet 298 00:17:30,400 --> 00:17:33,399 Speaker 2: could ever be totally free of ads, especially if the 299 00:17:33,440 --> 00:17:36,359 Speaker 2: content is being provided for free. I mean, somebody has 300 00:17:36,400 --> 00:17:39,200 Speaker 2: to get paid, right. So a few years ago, Ethan 301 00:17:39,280 --> 00:17:41,840 Speaker 2: and a colleague came up with a new proposal. If 302 00:17:41,880 --> 00:17:43,560 Speaker 2: ads are going to be on the Internet, and they're 303 00:17:43,560 --> 00:17:46,720 Speaker 2: crucial for it to function, could we make them somehow 304 00:17:46,840 --> 00:17:47,520 Speaker 2: more ethical. 305 00:17:47,880 --> 00:17:49,560 Speaker 1: And I'm going to get very personal about this for 306 00:17:49,600 --> 00:17:51,520 Speaker 1: a moment, because it's the example that we actually use 307 00:17:51,560 --> 00:17:54,720 Speaker 1: in the paper. I am an alcoholic. I am nine 308 00:17:54,800 --> 00:17:58,879 Speaker 1: years sober, but well through my twenties, thirties, forties, I 309 00:17:59,000 --> 00:18:02,920 Speaker 1: drank all and I was one of those drinkers where 310 00:18:02,920 --> 00:18:05,200 Speaker 1: it was like part of my personality, Like I collected 311 00:18:05,280 --> 00:18:11,159 Speaker 1: scotches and you know, I researched the interesting bourbons, and 312 00:18:11,200 --> 00:18:15,840 Speaker 1: so there's a whole digital trail associated with me as 313 00:18:15,920 --> 00:18:20,760 Speaker 1: a drinker. I go on to Twitter, and Twitter loves 314 00:18:20,960 --> 00:18:25,760 Speaker 1: advertising booze to me. They love putting up ads trying 315 00:18:25,760 --> 00:18:28,600 Speaker 1: to get me to buy a nice, high end bourbon. 316 00:18:29,640 --> 00:18:35,120 Speaker 1: I can't escape that data doppelganger. I can't escape that 317 00:18:35,600 --> 00:18:39,720 Speaker 1: other drunk version of myself, even though I'm now off 318 00:18:39,760 --> 00:18:42,480 Speaker 1: into the future. What we end up saying is that 319 00:18:43,040 --> 00:18:45,760 Speaker 1: you do have a right to be forgotten, not in 320 00:18:45,800 --> 00:18:48,040 Speaker 1: the sense of like you have a right necessarily be 321 00:18:48,119 --> 00:18:50,280 Speaker 1: scrubbed off the Internet. But you have a right to 322 00:18:50,359 --> 00:18:52,760 Speaker 1: change your mind. And so what we end up proposing 323 00:18:52,800 --> 00:18:57,840 Speaker 1: is forgetful advertising, and forgetful advertising basically says you can 324 00:18:57,960 --> 00:19:00,560 Speaker 1: know anything about me in this one session. You can 325 00:19:00,600 --> 00:19:02,119 Speaker 1: look at the different pages that I go to on 326 00:19:02,160 --> 00:19:04,359 Speaker 1: the website. You can look at the time, you can 327 00:19:04,400 --> 00:19:05,880 Speaker 1: look at the date, you can look at the location, 328 00:19:06,000 --> 00:19:08,040 Speaker 1: you can look at what I'm searching for. You can 329 00:19:08,080 --> 00:19:11,800 Speaker 1: look at the context of those pages and use that 330 00:19:11,840 --> 00:19:15,800 Speaker 1: to target an ad. What you cannot do is look 331 00:19:15,880 --> 00:19:18,000 Speaker 1: at what I was looking at a month ago, a 332 00:19:18,080 --> 00:19:23,120 Speaker 1: year ago, ten years ago. And we think that, yeah, 333 00:19:23,280 --> 00:19:26,720 Speaker 1: it'll be less accurate, but we think it's a more 334 00:19:26,760 --> 00:19:29,080 Speaker 1: ethical way to do it. And we end up suggesting 335 00:19:29,080 --> 00:19:31,920 Speaker 1: that it's something like free trade coffee. There are some 336 00:19:32,119 --> 00:19:35,600 Speaker 1: percentage of people who are willing to pay the money 337 00:19:36,119 --> 00:19:39,359 Speaker 1: to make sure that human beings aren't being exploited, you know, 338 00:19:39,400 --> 00:19:42,480 Speaker 1: for the product that we're drinking every morning. And so 339 00:19:42,560 --> 00:19:46,080 Speaker 1: what we were thinking was maybe there will be websites 340 00:19:46,119 --> 00:19:48,439 Speaker 1: that are willing to say we're going to do this. 341 00:19:48,880 --> 00:19:52,359 Speaker 1: I'll tell you my first attempt on this was The Guardian, 342 00:19:52,600 --> 00:19:58,760 Speaker 1: the British newspaper. They run all of these expos's on 343 00:19:58,760 --> 00:20:03,600 Speaker 1: online surveillance, they run all the stuff about Edward Snowden, 344 00:20:04,080 --> 00:20:06,760 Speaker 1: and I end up in a session with the publisher there, 345 00:20:06,800 --> 00:20:10,719 Speaker 1: Alan Russbritcher, and I say, Alan, I'm looking at these 346 00:20:10,720 --> 00:20:14,600 Speaker 1: stories on the Guardian and here are sixty trackers that 347 00:20:14,640 --> 00:20:17,439 Speaker 1: are sending the information. Then I'm looking at this on 348 00:20:17,480 --> 00:20:20,000 Speaker 1: the Guardian like it's time for you to get rid 349 00:20:20,000 --> 00:20:22,639 Speaker 1: of these things. You reading the Edward Snowden article and 350 00:20:22,680 --> 00:20:26,200 Speaker 1: there's trackers again. Yeah, but just the irony of that, 351 00:20:26,640 --> 00:20:29,119 Speaker 1: Like it feels like if you're going to be The 352 00:20:29,119 --> 00:20:33,720 Speaker 1: Guardian and a validly progressive publication, you might decide I 353 00:20:33,840 --> 00:20:36,080 Speaker 1: might opt out of this form of advertising and into 354 00:20:36,080 --> 00:20:37,600 Speaker 1: a more ethical form of advertising. 355 00:20:38,160 --> 00:20:42,600 Speaker 2: Okay, so I just want to echo the last thing 356 00:20:42,680 --> 00:20:46,040 Speaker 2: you said, ethical form of advertising. Might have remind you 357 00:20:46,119 --> 00:20:46,920 Speaker 2: that those are not. 358 00:20:47,040 --> 00:20:48,639 Speaker 1: Where it's the go together very often. 359 00:20:48,840 --> 00:20:51,600 Speaker 2: Yeah, ethical and advertising. I can't think of the last 360 00:20:51,600 --> 00:20:52,760 Speaker 2: time somebody said that to me. 361 00:20:53,119 --> 00:20:55,320 Speaker 1: Let's explore it for half a second, right, So let's 362 00:20:55,359 --> 00:20:58,879 Speaker 1: go and be Google for a bit. Okay, it's the 363 00:20:58,880 --> 00:21:03,920 Speaker 1: middle of winter here in western Massachusetts. My roof is leaking, 364 00:21:04,000 --> 00:21:08,040 Speaker 1: because everybody's roof is leaking. I go onto Google and 365 00:21:08,280 --> 00:21:13,040 Speaker 1: I search roof repair Pittsfield, Massachusetts. There are a whole 366 00:21:13,119 --> 00:21:17,400 Speaker 1: bunch of roofing contractors who really want my phone call, 367 00:21:18,080 --> 00:21:19,840 Speaker 1: and they want my phone call because so I call 368 00:21:19,880 --> 00:21:22,480 Speaker 1: them up. There's a decent chance that they're going to 369 00:21:22,520 --> 00:21:26,840 Speaker 1: make five to ten thousand dollars fixing my roof. That 370 00:21:27,040 --> 00:21:32,080 Speaker 1: ad click is worth anywhere between five and fifty bucks 371 00:21:32,520 --> 00:21:36,960 Speaker 1: in cost per customer acquisition. Google loves that ad click, 372 00:21:37,160 --> 00:21:40,119 Speaker 1: and frankly, some of the advertisers, the advertisers are like, man, 373 00:21:40,240 --> 00:21:44,520 Speaker 1: here is a qualified lead outside of Pittsfield, Massachusetts, to 374 00:21:44,600 --> 00:21:48,960 Speaker 1: go repair a roof. That's ethical advertising, right, That is 375 00:21:49,000 --> 00:21:53,680 Speaker 1: advertising that actually helps both me, somebody looking for a service, 376 00:21:54,200 --> 00:21:57,520 Speaker 1: someone trying to provide a service, and the broker in 377 00:21:57,560 --> 00:22:00,520 Speaker 1: the middle of those things. It is surveillance. It is 378 00:22:00,600 --> 00:22:04,119 Speaker 1: aware that I am near Pittsfield, Massachusetts, so there is 379 00:22:04,119 --> 00:22:07,440 Speaker 1: a certain amount of surveillance, but it's actually meeting a need. 380 00:22:10,080 --> 00:22:13,160 Speaker 2: In their paper on Forgetful Advertising, Ethan and his colleague 381 00:22:13,160 --> 00:22:17,240 Speaker 2: talk about what they call a DPI, a digital public infrastructure. 382 00:22:17,520 --> 00:22:20,160 Speaker 2: So think of it like this, Facebook and Google ad 383 00:22:20,200 --> 00:22:22,480 Speaker 2: Sense might just started out as little tools that could 384 00:22:22,480 --> 00:22:24,640 Speaker 2: help you talk to your college friends or maybe help 385 00:22:24,680 --> 00:22:27,159 Speaker 2: you sell t shirts. But nowadays, if you want to 386 00:22:27,200 --> 00:22:31,160 Speaker 2: sell anything online you have to use those platforms. There 387 00:22:31,200 --> 00:22:34,160 Speaker 2: really is no alternative. I mean, imagine if the only 388 00:22:34,200 --> 00:22:37,119 Speaker 2: way to drive from Santa Monica to downtown Los Angeles 389 00:22:37,320 --> 00:22:39,320 Speaker 2: was to take a toll road. You can't bike, you 390 00:22:39,320 --> 00:22:41,360 Speaker 2: can't walk, you can't take a bus, you can't take 391 00:22:41,359 --> 00:22:44,240 Speaker 2: a train, You can only take that expensive toll road. 392 00:22:44,720 --> 00:22:51,119 Speaker 2: There should be an alternative Forgetful advertising is just alternative infrastructure. Again, 393 00:22:51,119 --> 00:22:54,000 Speaker 2: they're not saying that they want to abolish Google ad Sense, 394 00:22:54,200 --> 00:22:57,479 Speaker 2: They're just building another option for people who prefer something, 395 00:22:57,680 --> 00:23:02,880 Speaker 2: as Ethan would put it, more ethical. But Ethan's work 396 00:23:02,920 --> 00:23:07,000 Speaker 2: on fixing the Internet doesn't stop there. He's got another idea, 397 00:23:07,080 --> 00:23:09,560 Speaker 2: and believe it or not, it kind of goes back 398 00:23:09,600 --> 00:23:11,920 Speaker 2: to that hack that he's most famous for. 399 00:23:12,800 --> 00:23:16,760 Speaker 1: Here's the interesting thing. Unless you are hanging out in 400 00:23:16,840 --> 00:23:19,919 Speaker 1: certain corners of the Internet and man that's between you 401 00:23:19,960 --> 00:23:23,680 Speaker 1: and your browser, you are probably not seeing a lot 402 00:23:23,720 --> 00:23:25,720 Speaker 1: of pop up ads these stays. And there's a couple 403 00:23:25,760 --> 00:23:29,800 Speaker 1: of reasons behind that. The first one is that every 404 00:23:30,080 --> 00:23:34,320 Speaker 1: modern web browser has a feature to allow you to 405 00:23:34,359 --> 00:23:40,320 Speaker 1: block pop up ads, and that is the simplest example 406 00:23:40,600 --> 00:23:45,600 Speaker 1: of middleware that we know of. So middleware is software 407 00:23:45,960 --> 00:23:50,520 Speaker 1: that you, as a user can install on your web 408 00:23:50,520 --> 00:23:54,200 Speaker 1: browser or potentially on your phone to give you more 409 00:23:54,440 --> 00:23:59,160 Speaker 1: control over a service. So the classic piece of middleware 410 00:23:59,640 --> 00:24:03,120 Speaker 1: is a pop up blocker, which is, hey, this site 411 00:24:03,160 --> 00:24:06,439 Speaker 1: wants to open a window. You can choose not to 412 00:24:06,520 --> 00:24:10,280 Speaker 1: have that opened if you prefer. So, first of all, 413 00:24:10,440 --> 00:24:14,000 Speaker 1: it actually shows us that things can get better. We're 414 00:24:14,040 --> 00:24:16,399 Speaker 1: not so much drowning and pop ups. We're now drowning 415 00:24:16,400 --> 00:24:19,520 Speaker 1: and other stuff. We're drowning in video AutoPlay ads. Sites 416 00:24:19,560 --> 00:24:26,159 Speaker 1: like Facebook work very actively to disable ad blockers, but 417 00:24:26,720 --> 00:24:31,160 Speaker 1: pop ups and ads in general were really the first 418 00:24:31,200 --> 00:24:33,399 Speaker 1: time that users were sort of like, you know what, 419 00:24:34,119 --> 00:24:37,119 Speaker 1: we actually need some more control in this situation. So 420 00:24:37,320 --> 00:24:40,240 Speaker 1: let's start building the software that gives us increasing amounts 421 00:24:40,240 --> 00:24:40,679 Speaker 1: of control. 422 00:24:42,680 --> 00:24:45,439 Speaker 2: And even once you use this concept of middleware to 423 00:24:45,480 --> 00:25:00,879 Speaker 2: fix social media, we'll explain how after the break. So 424 00:25:01,680 --> 00:25:04,280 Speaker 2: I just want to read back to you a headline 425 00:25:04,280 --> 00:25:06,960 Speaker 2: of an article that you wrote in twenty twenty four. 426 00:25:08,320 --> 00:25:13,960 Speaker 2: The headline reads, I love Facebook. That's why I'm suing Meta. Yeah, either, 427 00:25:13,960 --> 00:25:16,760 Speaker 2: please explain this to me. I guess I can come 428 00:25:17,040 --> 00:25:19,119 Speaker 2: across as kind of aggressive every so often. 429 00:25:19,640 --> 00:25:22,960 Speaker 1: So I'm a Facebook user. I use it pretty much 430 00:25:22,960 --> 00:25:27,840 Speaker 1: every day. I got married thanks to Facebook. Tenish years ago. 431 00:25:27,960 --> 00:25:32,080 Speaker 1: I was divorced. I reconnected with a college friend of mine. 432 00:25:32,160 --> 00:25:35,040 Speaker 1: She was living in Houston, Texas. I was out in 433 00:25:35,080 --> 00:25:37,960 Speaker 1: western Massachusetts. But I looked her up on Facebook and 434 00:25:38,000 --> 00:25:40,240 Speaker 1: we started chatting and I invited her to talk that 435 00:25:40,280 --> 00:25:43,040 Speaker 1: I was giving down at Rice University. Seven years later, 436 00:25:43,080 --> 00:25:47,160 Speaker 1: we're married. So thank you Facebook. You know, I appreciate that, 437 00:25:47,240 --> 00:25:51,639 Speaker 1: and I think as far as keeping people loosely in 438 00:25:51,720 --> 00:25:53,879 Speaker 1: touch with one another, particularly you know, for me as 439 00:25:53,920 --> 00:25:59,400 Speaker 1: a gen xer, it's actually been an incredible tool. However, Comma, 440 00:26:00,359 --> 00:26:04,840 Speaker 1: like everything else, it is becoming more and more algorithmically driven. 441 00:26:05,440 --> 00:26:08,439 Speaker 1: And these days on Facebook, I am a lot less 442 00:26:09,119 --> 00:26:12,479 Speaker 1: likely to be seeing updates from my friends about their kids. 443 00:26:13,280 --> 00:26:17,119 Speaker 1: I instead get a lot of reels just trying to 444 00:26:17,119 --> 00:26:19,960 Speaker 1: capture as much of my attention as possible. And so 445 00:26:20,560 --> 00:26:27,080 Speaker 1: I got really interested in this idea that maybe I 446 00:26:27,119 --> 00:26:30,199 Speaker 1: could have more control over Facebook. Friend of mine in 447 00:26:30,200 --> 00:26:33,800 Speaker 1: the UK, guy named Louis Barclay, wrote a piece of 448 00:26:33,920 --> 00:26:38,439 Speaker 1: software called unfollow everything. It did something super simple. It 449 00:26:38,520 --> 00:26:41,879 Speaker 1: went into Facebook. It kept your whole friend's graph. It 450 00:26:41,960 --> 00:26:43,879 Speaker 1: just said I'm going to unfollow all of them, and 451 00:26:43,960 --> 00:26:47,520 Speaker 1: eventually what you end up with is an empty feed. 452 00:26:48,000 --> 00:26:51,359 Speaker 1: You just don't have anyone that you're following anymore. People 453 00:26:51,440 --> 00:26:54,119 Speaker 1: liked it, Louie liked it, other people started using it, 454 00:26:54,320 --> 00:26:57,400 Speaker 1: and Facebook said, Louis, you got to take this down 455 00:26:57,840 --> 00:27:00,600 Speaker 1: otherwise we're going to come after you and you court. 456 00:27:01,160 --> 00:27:05,280 Speaker 1: And by the way, that long click rap agreement that 457 00:27:05,320 --> 00:27:11,280 Speaker 1: you signed onto makes you responsible for our legal fees. 458 00:27:12,040 --> 00:27:17,160 Speaker 1: So what Facebook's contract in the UK says that if 459 00:27:17,160 --> 00:27:20,840 Speaker 1: you lose litigation against them, you are responsible for the 460 00:27:20,920 --> 00:27:25,159 Speaker 1: legal fees incurred by Facebook. Wow. So Louis looks at 461 00:27:25,160 --> 00:27:29,000 Speaker 1: this and says, yeah, I'm not taking that gamble. I 462 00:27:29,080 --> 00:27:32,840 Speaker 1: could be financially ruined. But he goes to a group 463 00:27:32,880 --> 00:27:36,400 Speaker 1: called the Night First Amendment Institute at Columbia University and 464 00:27:36,440 --> 00:27:39,879 Speaker 1: they say, what if we tried to do this again 465 00:27:40,320 --> 00:27:42,399 Speaker 1: in the United States? And I said, you know, I 466 00:27:42,520 --> 00:27:46,720 Speaker 1: love this idea because this is an example of middleware. 467 00:27:47,600 --> 00:27:50,320 Speaker 1: So Night First Amendments that comes to me and says, 468 00:27:51,320 --> 00:27:53,399 Speaker 1: do you want to try to put this in front 469 00:27:53,400 --> 00:27:57,320 Speaker 1: of a federal court, and the idea is, I will 470 00:27:57,320 --> 00:28:00,720 Speaker 1: put together an experiment. The experiment is we will recruit 471 00:28:00,720 --> 00:28:04,679 Speaker 1: a whole bunch of users, they will try using an 472 00:28:04,720 --> 00:28:08,439 Speaker 1: updated version of unfollow Everything, and they will let us 473 00:28:08,480 --> 00:28:12,080 Speaker 1: know whether they like that experience, and we will study 474 00:28:12,680 --> 00:28:15,720 Speaker 1: whether it changes how much they use Facebook. We then 475 00:28:15,800 --> 00:28:18,840 Speaker 1: brought it to a federal court and said, hey, we 476 00:28:18,920 --> 00:28:22,480 Speaker 1: want to do this. The last time someone did this, 477 00:28:22,960 --> 00:28:25,000 Speaker 1: Facebook came down on them like a ton of bricks. 478 00:28:25,600 --> 00:28:28,600 Speaker 1: Before we invest our time and money on this, we'd 479 00:28:28,640 --> 00:28:32,320 Speaker 1: like a ruling that this is legal. The federal court said, nice, 480 00:28:32,359 --> 00:28:36,679 Speaker 1: tryprof You're asking us to rule on something that hasn't 481 00:28:36,680 --> 00:28:39,520 Speaker 1: happened yet. That's not how the US court system works. 482 00:28:40,480 --> 00:28:44,000 Speaker 1: Build something and we'll let you know what happens. So 483 00:28:44,200 --> 00:28:45,800 Speaker 1: I have now been in the process of trying to 484 00:28:45,800 --> 00:28:48,920 Speaker 1: figure out how to build what my friend Louie built. 485 00:28:49,640 --> 00:28:53,840 Speaker 1: The trick is in the couple of years since Louie 486 00:28:53,840 --> 00:28:58,200 Speaker 1: built that software, Facebook has put a lot of people 487 00:28:59,040 --> 00:29:04,000 Speaker 1: working on making middlewhere impossible. Specifically, the middleware that they 488 00:29:04,120 --> 00:29:07,800 Speaker 1: care about is ad blockers, and they have a whole 489 00:29:07,880 --> 00:29:11,800 Speaker 1: team working to make it almost impossible to block ads. 490 00:29:11,800 --> 00:29:15,280 Speaker 1: On Facebook, and the same techniques that I need to 491 00:29:15,440 --> 00:29:18,440 Speaker 1: use to do and follow everything are the same techniques 492 00:29:18,440 --> 00:29:23,080 Speaker 1: that ad blockers use. So we're still very much engaged 493 00:29:23,240 --> 00:29:28,240 Speaker 1: in battle. But it's me and a couple of brave 494 00:29:28,320 --> 00:29:34,200 Speaker 1: students that ums Amherst versus Facebook, all of its resources 495 00:29:34,760 --> 00:29:36,160 Speaker 1: and all of its lawyers. 496 00:29:37,400 --> 00:29:39,520 Speaker 2: I don't know too many people who are willing to 497 00:29:39,560 --> 00:29:42,320 Speaker 2: go against Meta like that, but if this thing that 498 00:29:42,400 --> 00:29:45,160 Speaker 2: Ethan is working on comes out, I would try it. 499 00:29:45,240 --> 00:29:46,720 Speaker 2: I mean, this would be the one thing that could 500 00:29:46,720 --> 00:29:50,160 Speaker 2: bring me back to Facebook. But Ethan isn't advocating for 501 00:29:50,200 --> 00:29:51,960 Speaker 2: everyone to just sit back and wait for him to 502 00:29:51,960 --> 00:29:54,200 Speaker 2: swoop in and save the day. There are things that 503 00:29:54,240 --> 00:29:57,240 Speaker 2: you and me can do right now for ourselves. It's 504 00:29:57,280 --> 00:29:58,240 Speaker 2: just going to take some work. 505 00:29:59,200 --> 00:30:02,320 Speaker 1: So the funny thing is you can do things about 506 00:30:02,360 --> 00:30:04,320 Speaker 1: it right So here's a couple of things you can 507 00:30:04,360 --> 00:30:08,400 Speaker 1: do about it. A lot of people have gotten off 508 00:30:08,440 --> 00:30:13,880 Speaker 1: of Twitter because Twitter is now controlled by a right 509 00:30:13,920 --> 00:30:17,280 Speaker 1: wing extremist troll and it is not the community that. 510 00:30:17,240 --> 00:30:17,800 Speaker 3: It used to be. 511 00:30:18,360 --> 00:30:21,240 Speaker 1: When people get off Twitter, they tend to go one 512 00:30:21,240 --> 00:30:24,280 Speaker 1: of two different places, and they actually have both really 513 00:30:24,280 --> 00:30:27,680 Speaker 1: good things to talk about with them. Blue Sky has 514 00:30:27,720 --> 00:30:30,480 Speaker 1: a lot of the functionality that Twitter used to have, 515 00:30:30,800 --> 00:30:34,360 Speaker 1: and it's got some very cool functionality around being able 516 00:30:34,360 --> 00:30:37,720 Speaker 1: to do custom feeds. You can go into Blue Sky 517 00:30:37,800 --> 00:30:39,520 Speaker 1: and you can actually have a lot of control, very 518 00:30:39,600 --> 00:30:43,000 Speaker 1: much like middleware, except that it's actually built into the tool. 519 00:30:43,120 --> 00:30:47,000 Speaker 1: It's built in as a function. People are using their 520 00:30:47,040 --> 00:30:50,360 Speaker 1: ability to do feeds to do very focused collections. There's 521 00:30:50,360 --> 00:30:52,760 Speaker 1: a collection called Black Sky for people who are trying 522 00:30:52,800 --> 00:30:56,239 Speaker 1: to get more color in their Twitter feeds. There are 523 00:30:56,320 --> 00:30:59,120 Speaker 1: people doing it to push women, to push voices from 524 00:30:59,160 --> 00:31:03,520 Speaker 1: the developing world. It's a really exciting set of tools 525 00:31:04,000 --> 00:31:06,320 Speaker 1: and I'm a huge supporter of it. The trick is 526 00:31:06,360 --> 00:31:09,680 Speaker 1: Blue Sky is still run by a corporation. They're subject 527 00:31:09,680 --> 00:31:11,520 Speaker 1: to the laws of capitalism, so on and so forth. 528 00:31:11,840 --> 00:31:15,800 Speaker 1: So let me introduce you to mastadon. Mastodon is an 529 00:31:15,920 --> 00:31:19,760 Speaker 1: open source alternative to Twitter. You can run a mast 530 00:31:19,800 --> 00:31:22,520 Speaker 1: it on server on your own server. You probably don't 531 00:31:22,560 --> 00:31:25,440 Speaker 1: want to. It's a decent amount of work, but I 532 00:31:25,520 --> 00:31:28,320 Speaker 1: am part of a group called social dot co Op. 533 00:31:29,040 --> 00:31:31,720 Speaker 1: This is a masked it on server that is run 534 00:31:31,840 --> 00:31:34,160 Speaker 1: as a co op. I pay seventy five bucks a 535 00:31:34,240 --> 00:31:37,600 Speaker 1: year for my membership to it. The leadership of it, 536 00:31:37,640 --> 00:31:41,320 Speaker 1: the people who do moderation. It is elected out of 537 00:31:41,320 --> 00:31:43,920 Speaker 1: the community. People are asked to put in volunteer time 538 00:31:44,000 --> 00:31:47,040 Speaker 1: to take on moderation duties at different points. It is 539 00:31:47,080 --> 00:31:51,400 Speaker 1: a different model for doing this. The trick is these 540 00:31:51,560 --> 00:31:56,280 Speaker 1: models are small. They're getting used by a couple million people, 541 00:31:56,680 --> 00:32:01,560 Speaker 1: not by two billion people. These models don't have ad budgets. 542 00:32:01,600 --> 00:32:06,400 Speaker 1: They're not going out and buying Super Bowl ads. And 543 00:32:07,400 --> 00:32:09,760 Speaker 1: we got to tell the story. We got to get 544 00:32:09,760 --> 00:32:13,719 Speaker 1: on podcasts and tell people that there's another way to 545 00:32:13,840 --> 00:32:17,239 Speaker 1: do this. It doesn't have to suck as bad as 546 00:32:17,280 --> 00:32:20,600 Speaker 1: it does. It really doesn't. So here are three things 547 00:32:20,600 --> 00:32:24,520 Speaker 1: that we can do. We need to force these big 548 00:32:24,800 --> 00:32:30,360 Speaker 1: platforms to allow for middlework. We need to make it 549 00:32:30,760 --> 00:32:36,320 Speaker 1: a law that you can use something to give you 550 00:32:36,360 --> 00:32:41,000 Speaker 1: more control over TikTok or Facebook or something along those lines. Second, 551 00:32:41,880 --> 00:32:45,840 Speaker 1: we would do interoperability, so if you don't want to 552 00:32:45,960 --> 00:32:49,400 Speaker 1: use TikTok's app to see TikTok, you can still get 553 00:32:49,440 --> 00:32:51,360 Speaker 1: the content out of it, but you can do it 554 00:32:51,480 --> 00:32:55,920 Speaker 1: through a browser on your phone that you control. And 555 00:32:55,960 --> 00:33:01,520 Speaker 1: then third, we got to get people psyched about building 556 00:33:01,680 --> 00:33:07,720 Speaker 1: and governing their own online spaces. Right now, social dot 557 00:33:07,760 --> 00:33:12,200 Speaker 1: co op is not a good alternative to Twitter or 558 00:33:12,200 --> 00:33:14,720 Speaker 1: to blue Sky. There just aren't enough people on mastadon. 559 00:33:15,480 --> 00:33:18,520 Speaker 1: But if I have a client on my phone that 560 00:33:18,560 --> 00:33:21,560 Speaker 1: I control and I can merge them all together, Hey, 561 00:33:21,760 --> 00:33:23,560 Speaker 1: let me see what people are talking about on Twitter, 562 00:33:23,640 --> 00:33:25,960 Speaker 1: let me see Facebook, let me see Insta, let me 563 00:33:26,000 --> 00:33:28,640 Speaker 1: see all of it and my own little social network 564 00:33:28,640 --> 00:33:31,800 Speaker 1: at the same place. And I can go in and say, 565 00:33:31,840 --> 00:33:34,720 Speaker 1: you know what, and prioritize my local news and you 566 00:33:34,760 --> 00:33:38,800 Speaker 1: know what, I'm probably not hearing enough from women in 567 00:33:38,840 --> 00:33:41,719 Speaker 1: the global South, So you know, adding a feed like that, 568 00:33:42,520 --> 00:33:46,240 Speaker 1: now I've got my custom feed. It's local, it's prioritizing 569 00:33:46,240 --> 00:33:48,640 Speaker 1: the voices I want to hear, it's incorporating all of 570 00:33:48,640 --> 00:33:53,440 Speaker 1: these things. This is technically possible. We have shown we 571 00:33:53,520 --> 00:33:57,400 Speaker 1: can do it. It is a policy issue, and it 572 00:33:57,480 --> 00:34:00,640 Speaker 1: is a policy issue that's very hard to make progress 573 00:34:00,640 --> 00:34:04,960 Speaker 1: on in the United States because these very powerful platforms 574 00:34:05,160 --> 00:34:09,279 Speaker 1: are run by men who are sitting with Donald Trump 575 00:34:09,360 --> 00:34:16,279 Speaker 1: at his inauguration period. These companies figured out that they 576 00:34:16,520 --> 00:34:19,080 Speaker 1: needed to stay on the right side of the most 577 00:34:19,120 --> 00:34:22,520 Speaker 1: powerful person in the United States right now, so that 578 00:34:22,560 --> 00:34:24,080 Speaker 1: we don't get a better internet. 579 00:34:24,800 --> 00:34:27,560 Speaker 2: What you're talking about is I mean even the phrase 580 00:34:27,600 --> 00:34:29,920 Speaker 2: a better Internet. I mean, I'm sure you've seen that 581 00:34:30,200 --> 00:34:32,840 Speaker 2: a lot of people's solution is just get off of 582 00:34:32,880 --> 00:34:35,840 Speaker 2: social media. It's really interesting to me that you're actually 583 00:34:35,880 --> 00:34:41,880 Speaker 2: advocating for in some ways basically saving social media for 584 00:34:42,040 --> 00:34:43,879 Speaker 2: people who are trying to leave it. 585 00:34:44,560 --> 00:34:48,360 Speaker 1: Social media is worth saving. I'm a Type one diabetic. 586 00:34:48,520 --> 00:34:51,239 Speaker 1: I'm an insulin dependent diabetic. I've had diabetes since I 587 00:34:51,280 --> 00:34:54,200 Speaker 1: was thirteen years old. I'm fifty three years old. Now 588 00:34:54,680 --> 00:34:57,400 Speaker 1: my body's changing. I'm trying to figure out how to 589 00:34:57,440 --> 00:35:02,960 Speaker 1: manage my diabetes, lose way, keep up my running. And 590 00:35:03,440 --> 00:35:05,160 Speaker 1: one of the things that I rely on to do 591 00:35:05,200 --> 00:35:08,759 Speaker 1: this is the t one D community on Reddit, where 592 00:35:08,760 --> 00:35:11,319 Speaker 1: I can come on and say, hey, I go for 593 00:35:11,400 --> 00:35:13,719 Speaker 1: runs in the morning and my blood sugar goes up. 594 00:35:13,800 --> 00:35:15,839 Speaker 1: What's up with that? And someone comes in and says, 595 00:35:15,880 --> 00:35:18,600 Speaker 1: it's called a cortisol dump. Here's the scientific paper on it. 596 00:35:18,640 --> 00:35:21,360 Speaker 1: Find out more about it. That is the Internet that 597 00:35:21,400 --> 00:35:24,600 Speaker 1: I fell in love with in nineteen eighty nine when 598 00:35:24,640 --> 00:35:27,799 Speaker 1: I showed up at my university and discovered that I 599 00:35:27,840 --> 00:35:31,680 Speaker 1: could log onto incredible amounts of information out there. The 600 00:35:31,760 --> 00:35:34,839 Speaker 1: Internet I love is the Internet that I was able 601 00:35:34,920 --> 00:35:38,120 Speaker 1: to turn to and say, hey, I wonder where Amy 602 00:35:38,239 --> 00:35:42,319 Speaker 1: is these days, and reconnect with her and seven years 603 00:35:42,360 --> 00:35:47,200 Speaker 1: later get married. It is the internet where queer and 604 00:35:47,320 --> 00:35:51,560 Speaker 1: trans kids in my community can go online and find 605 00:35:51,600 --> 00:35:53,640 Speaker 1: people who are going through the same experiences that they're 606 00:35:53,680 --> 00:35:56,640 Speaker 1: going through in high school, find out about the summer 607 00:35:56,680 --> 00:35:59,000 Speaker 1: camps that they can go to, and they can bond 608 00:35:59,040 --> 00:36:02,799 Speaker 1: with other trans us. This is what social media should 609 00:36:02,840 --> 00:36:09,280 Speaker 1: be about. Social media doesn't have to be terrible, and 610 00:36:09,920 --> 00:36:15,040 Speaker 1: it is not our fault. It has become terrible because 611 00:36:15,080 --> 00:36:19,640 Speaker 1: some very powerful people with some very good financial motivations 612 00:36:20,160 --> 00:36:23,680 Speaker 1: have figured out how to make money off it being terrible. 613 00:36:24,120 --> 00:36:29,640 Speaker 1: But it will continue to be terrible until we take 614 00:36:29,719 --> 00:36:31,840 Speaker 1: responsibility for it not being terrible. 615 00:36:34,520 --> 00:36:38,160 Speaker 2: I have started to hate social media. I've basically stopped 616 00:36:38,200 --> 00:36:40,960 Speaker 2: using it, which has been pretty good for my mental health, 617 00:36:41,040 --> 00:36:43,360 Speaker 2: but it also means that it's much harder for me 618 00:36:43,400 --> 00:36:45,520 Speaker 2: to stay in tune what my parents or my other 619 00:36:45,560 --> 00:36:48,400 Speaker 2: family members are into. So what if we could just 620 00:36:48,560 --> 00:36:50,839 Speaker 2: plug something into social media and make it a little 621 00:36:50,840 --> 00:36:54,120 Speaker 2: bit better. What if Ethan's middleware project could make you say, 622 00:36:54,160 --> 00:36:58,000 Speaker 2: you know what, I can tolerate Facebook now It's probably 623 00:36:58,080 --> 00:37:00,680 Speaker 2: not the only solution, and it might not be for everyone, 624 00:37:01,120 --> 00:37:03,239 Speaker 2: but I do think that it's an option that, at 625 00:37:03,280 --> 00:37:06,480 Speaker 2: the very least we should be allowed to try. Anyway, 626 00:37:06,600 --> 00:37:08,520 Speaker 2: that's it for this one. And by the way, if 627 00:37:08,560 --> 00:37:10,560 Speaker 2: you like this episode and you want to know more 628 00:37:10,640 --> 00:37:13,000 Speaker 2: about why it is that the Internet got so bad, 629 00:37:13,360 --> 00:37:15,520 Speaker 2: check out my interview with Tim Woo. The link for that, 630 00:37:15,719 --> 00:37:18,160 Speaker 2: plus links to Ethan's work, is in the show notes. 631 00:37:22,680 --> 00:37:25,080 Speaker 2: Thank you again to Ethan for breaking all this down 632 00:37:25,120 --> 00:37:27,759 Speaker 2: for us, and thank you once again for listening to 633 00:37:27,800 --> 00:37:30,359 Speaker 2: another episode at kill Switch. If you want to talk, 634 00:37:30,400 --> 00:37:34,200 Speaker 2: you can email us at kill Switch at Kaleidoscope dot NYC, 635 00:37:34,680 --> 00:37:38,080 Speaker 2: or on Instagram. We're at kill switchpod. And if you 636 00:37:38,120 --> 00:37:40,120 Speaker 2: dug this one, I think someone else might like it too, 637 00:37:40,280 --> 00:37:42,080 Speaker 2: you could send it to them, or if you're too 638 00:37:42,120 --> 00:37:44,160 Speaker 2: shy to share us with your friends, you could write 639 00:37:44,200 --> 00:37:46,720 Speaker 2: a review. It helps other people find the show, which 640 00:37:46,760 --> 00:37:50,160 Speaker 2: helps us keep doing our thing. Kill Switch is hosted 641 00:37:50,160 --> 00:37:53,359 Speaker 2: by Me Dexter Thomas. It's produced by Shena Ozaki, dar 642 00:37:53,480 --> 00:37:56,600 Speaker 2: Luck Potts, and Julia Nutter. A theme song is by 643 00:37:56,640 --> 00:37:59,359 Speaker 2: me and Kyle Murdoch, and Kyle also mixed the show. 644 00:38:00,000 --> 00:38:03,200 Speaker 2: From Kaleidoscope. Our executive producers are Oswa Lashin One Guest 645 00:38:03,239 --> 00:38:06,920 Speaker 2: Shot to Gadur and Kate Osborne from iHeart. Our executive 646 00:38:06,920 --> 00:38:11,000 Speaker 2: producers are Katrina Norville and Nikki Detour. Catch on the 647 00:38:11,040 --> 00:38:21,600 Speaker 2: next one, Goodbye,