Today on Reimagining, we welcome our first conscientious objector to Google—and our first ever NASA alum. Janet Vertesi joins for a fascinating conversation about her project to keep any data about her children off the web, and ties it in to tales about her old job as in-house ethnographer for the Mars Rover missions.
Janet Vertesi is an Associate Professor of Sociology at Princeton University, and has published two books about her work with NASA: 2015’s Seeing Like a Rover: How Robots, Teams, and Images Craft Knowledge of Mars and 2020’s Shaping Science: Organizations, Decisions, and Culture on NASA’s Teams. She recently published a fascinating article in Public Books about trying to avoid data collection at Disney World.
We mention Arvind Narayanan in this interview, who we had on the show last year.
Transcript
Ethan Zuckerman:
Hey, everybody, welcome back to Reimagining the Internet. I remain your host, Ethan Zuckerman. Thrilled to be with today, sociologist of science and technology at Princeton University, where she is an associate professor in sociology: Janet Vertesi. Janet has written some fascinating stuff over the last few years, starting with a pair of books about the Mars Rover missions that she researched with National Science Foundation support. The most recent of the two came out in 2020 with University of Chicago Press called Shaping Science: Organizations, Decisions, and Culture on NASA’s Teams. But Janet is also very well known for a set of experiments with online privacy. She describes herself as a conscientious objector to Google, and she’s had some success in keeping aspects of her personal life off the internet, which felt like a great topic for us to talk about. Janet, welcome.
Janet Vertesi:
Thanks so much, Ethan. I’m really happy to be here.
Ethan Zuckerman:
One of the ways that I got to know your work was a wonderful experiment you engaged in in 2014, where you, with the cooperation of your husband and the help of the Tor browser, worked really hard to keep your pregnancy off the internet. Now this followed a story that got a great deal of attention about the corporation Target figuring out that a young woman was pregnant before she, in fact, figured out she was pregnant. What made you so determined to keep your pregnancy off the internet in 2014?
Janet Vertesi:
It’s a great question because certainly the target teenager was on our minds when we first started talking about it. I got married just before the Snowden revelations came out. I had gotten off Google a year earlier. I had been very concerned about what I was seeing in the space that we were calling big data and privacy. In particular, the kind of convergence of people’s data sets to produce insights about them, but also to target them for advertising. And the fact that I had learned that a pregnant woman’s data—once a pregnant woman has been detected, that data is so valuable. That targeting data is so valuable. It just moves, right? If you can sell that data at a top dollar to someone else who’s going to sell Pampers or Huggies to that woman, that woman has lost control over where their data goes. So I had just spent the better part of a year thinking very carefully about how and where I put my data after leaving Google and thinking very strategically about how to guard my privacy and data sovereignty, really my ability to control where my data went in alignment with my values and my relationships.
So the last thing I wanted was to put some announcements or some piece stick up on Facebook like everyone else was doing or an ultrasound picture and have like Mark Zuckerberg know that I was pregnant because as soon as that data was on Facebook, that would be the end. I’d also just planned a wedding where my husband and I were sitting there playing Starcraft too, every night, and coding a wedding website from scratch because we didn’t want to use Google. And he got all these cool gamer gear ads and I got like bridesmaids ads and yoga retreats. And I just thought, soon as the Internet knows, I’m just going to be like summer camps and diaper ads and like, that’s it. I’m just going to be the mom and I’m so much more than that.
So that’s really where it started was this sensibility towards data and data mobility, personal data mobility. And then try to figure out how to evade that.
Ethan Zuckerman:
In one of your op-eds, you report that the data of a pregnant woman can be worth 200 times the data of a non-pregnant person, presumably because you’re making a lot of decisions about where to shop and what brand of diapers, so on and so forth that lock you in as a valuable customer over time. But it sounds like your concern about private data online precedes that and you traced it back to a change of privacy policy at Google in 2012. What did Google do in 2012 that sent you off the Googles?
Janet Vertesi:
Oh man, Ethan, Google broke my heart in 2012. I mean, Google and I were in this like long-term intimate relationship. I had one of those sort of top-level Gmail addresses. I was in there super early. It ran my calendar, it knew which trains I took, it just did everything for me. And then the cardinal sin was Google decided that they were going to amalgamate your data across all of its various platforms and services. So until then, what happened on YouTube, stayed on YouTube, what happened in mail, stayed on mail, what you did on maps, stayed in maps, and never the thrain should meet, right?
Google said largely in response to a competition from Facebook, we’re going to push your user ID as that sole token around which we’re going to gather all this data. And we’re going to use that to show you really useful things. It’s going to be great. You’ll see really useful stuff. But I knew, you know, I’m sitting at Princeton next to like Arvind Narayanan who talks about the degrees of entropy, right? Two data sets together, three data sets together. of those data sets are anonymized, you’ve targeted an individual. You can pinpoint a person. And I’m just thinking, you know, what of the history of surveillance and data capture in humankind did you get so wrong, Google? And it’s not like they didn’t know. I mean, these were, this was a presidential commission on big data. These are people who, you know, Ed Felton went through the Federal Trade Commission. It’s not like they didn’t know. But they did it anyway.
And when that happened, I remember sitting on a subway in New York City. I was on my way to some big data conference. And there were these really friendly ads all over the subway car. And they were all in primary colors and cartoons. And they were like, “Google, care about your privacy. Make sure to keep things locked up. Don’t share your passwords. Oh, by the way, a privacy policy change is coming.” And I thought, I can’t stay. I can’t click accept. When this comes down the pipe, I know too much to say yes.
I tried writing them in email, there was nowhere to phone. And there wasn’t even a button to say no. I just had to say, OK, see you later.
Ethan Zuckerman:
So what’s that process like? Because I’ll freely admit here that of all the big scary tech companies out there, Google surely knows the most about me. I still do my email through Gmail. It has my calendar. Absolutely, it’s got me dead to rights. And so far, the utility and the ease of it has been a tradeoff that I’m about that only willing to make. Or I simply haven’t thought about it as thoroughly as you have. What’s the process like of de-Googling your life?
Janet Vertesi:
Well, the first thing you have to realize that you’re in a codependent relationship. You need Google, but Google needs you. And so it’s going to manipulate you into staying. And now there’s actually some publications about this, the kind of dark patterns of these tech companies engage in to try to keep you online. At the time, I was just kind of encountering this, you know, for the first time, like trying a breakup and Google says, “Are you sure you want to do this? We’ve shared 38,000 conversations since 2008.” And I’m like, “No, I feel so guilty.” They’re going to lay the guilt on and they’re going to make you feel like it’s so hard to go elsewhere.
Ethan Zuckerman:
What was there a moment where Google put the boom box above their head and played Peter Gabriel’s “In Your Eye3s? or does it just sort of…
Janet Vertesi:
Yeah basically, full on. There was a moment where I deleted all my mail and it literally said you don’t have any mail. Our servers are feeling unloved and then it had a sad face smiley. So that’s kind of the equivalent of “In Your Eyes.” Yeah. I mean, it was very much like come back to the table. We got this romantic dinner set for you.
Ethan Zuckerman:
Yeah, just email anything.
Janet Vertesi:
Right, yeah. I think I’d been through enough failed relationships in my 20s to recognize that when I saw it. And was just like, yeah, absolutely not. But the key thing, actually—and I’ve been thinking a lot more about this since the personal data economy is really expanded and since my own opt out experiments have expanded—is one of the ways they get you is through convenience. And it’s through—we live in a capitalist society in which everything we do is measured in some form of efficiency. And we are more likely to accept these trades for our personal data privacy in exchange for couple of seconds less commute, a little bit of money off at the till, not having to be your own system administrator. It’s kind of obnoxious at home. We’re willing to make those trades and the companies have learned that.
And so what they start doing is they’ve now created an entire minefield of what I call microconveniences that are just there to make you be like, yeah, yeah, I’ll let that data move. No problem because I’m getting something for it. I don’t have to copy and paste this address into a field. It’s doing it for me behind those scenes. I’m gonna conveniently ignore what that action fosters in terms of the problems of the personal data economy. In exchange for, I got three seconds back in my day. And the problem is, is not only do they create these microconveniences, they make it harder for you to find an alternative. So you’re sitting there going, but what would I use instead of Google? Oh my goodness, I would be lost.
Well, actually, you wouldn’t be lost. There are alternatives out there, but it’s in Google’s best interest to make sure you don’t find them. They don’t seem as user-friendly, or that they buy up their competition so that you never have to encounter them. And that’s the kind of stuff I’m really pushing back against in my opt out projects today.
Ethan Zuckerman:
But before we get into the opt out projects, which I want to look at, which are amazing and wonderful and worth looking at, sometimes it’s not three seconds, right? So I use Gmail. One of the reasons that I use Gmail is that I got really sick of administering my own mail server. Spam turns out to be a real thing, even though you don’t encounter it so much if you’re on a system like Gmail at this point, how are you wrestling with the cases where it’s not a matter of light conveyance, where it really is almost sort of a lifestyle choice.
Janet Vertesi:
It is a lifestyle choice, and you’re right, it does get increasingly inconvenient and a heavy load. I mean, just trying to watch a television show and get your Plex server to talk to where your movies are, and you’re, you know, I mean, it’s just, it is complicated. And then your, your Raspberry Pi that’s doing some crucial thing goes down or someone’s not upgrading the firmware and you’re kind of living in this life of perpetual tinkering. So it does become a lifestyle choice.
However, I think I embrace that because it shows that I’m in control. And then the other thing is I do it because, I don’t know if you’ve heard of this concept in sociology, it’s called the breaching experiment. And the idea is that you take the rules that everybody is currently doing about social life, and then you break them and you see what happens. So a classic case would be you’re sitting there playing chess with someone, they move a pawn, you move a pawn, and then you pick up your rook and take their queen, which is like an illegal move. And then you see what they do. A lot of people get very frustrated. But when you start doing this with the rules of social life, you’re actually able to peel back the layers of the onion and understand what’s actually going on there. And so for me, those moments become instructive. They’re breaching experiments that are showing what’s really happening under the surface.
One thing I’ve noticed from administering mail servers is not the problem with spam, like I’m getting spam. The problem with mail servers now is that the big tech companies have consolidated so much that they’re blocking everybody else’s mail as spam. So I’m trying to send mail from my own server, and it’s getting spam boxed and no one can see it. ‘Cause they’re putting in a whole series of new rules that allow them to play by their game, and not anybody else’s. I wouldn’t be able to see that that’s going on and be able to articulate that, why it’s happening or fight against that if I weren’t taking this chance I’m taking.
Ethan Zuckerman:
And that is one of those moments where this sort of possibility of the Internet as a whole bunch of independent but under connected systems starts disappearing, right? I mean, very fundamental to how we think of the Internet is this idea that it is distributed, not just decentralized, that it’s all of these points very loosely joined to one another. But in a very practical way, email isn’t all that much anymore. And if you’re not getting routed by one of those really big mail servers, it becomes very challenging.
Going back to that breaching experiment, have you had reactions where people look at somebody like the opt out experiments that you’ve been engaged in and respond with anger, respond with a certain amount of outrage associated with it?
Janet Vertesi:
Oh my gosh, all the time. I’ll never forget at this one of the early meetings I had about the digital humanities initiative here at Princeton and one of the guys turned to me and he was like, “You’re not still doing that, like no Google thing, are you?” I said, “Are you gonna make it hard for us to share documents?” And I felt like, if you’re a vegetarian and you go home for Thanksgiving and your mom’s, like, you’re not still doing that vegetarian thing, are you? And I thought, yeah, absolutely, many of them get upset because you’re the fly in the ointment, like you’re the one that’s making it not easy for them. And they’re using Google because it’s easy for them too.
That said, I get a lot of people now, increasingly I get people who want to know how I did it. and they’re excited about it. And they’re like, “Yeah, I’ll help you with that. No problem. Even if I have to email you this file every week independently so that you don’t have to touch a Google site, I’m still gonna do it ’cause I support that this is the thing that you’re doing.” So yeah, initially it causes a lot of anger for some people it causes a lot of anger. And for others, they get excited about joining the resistance.
Ethan Zuckerman:
So I wanna talk about the full dynamics of the resistance. I assume that once you got off of Google, you got rid of your Android phone and switch right over to Apple, right?
Janet Vertesi:
Yeah, I never actually had an Android phone. I did have an iPhone. And then actually just after having my first kid, I just after having my first kid, I started looking differently at the cell phone. Because at that point, cell phones were changing too. And by 2014, I was like, this, everything was moving to the cloud. The clouds were all owned by various companies. The cloud is just someone else’s computer, right? I mean, yeah.
Ethan Zuckerman:
It’s the computer that you don’t have sudo on, yeah.
Janet Vertesi:
Exactly. I’m not root there. So I don’t know what they’re doing with it. So I didn’t want to participate in that. And I actually experimented with building my own cell phone, which it turns out is not the greatest idea. I mean, it worked kind of, but it didn’t do the things I actually needed a smartphone to do.
Ethan Zuckerman:
And when you say build your own, do you mean sort of start with a Linux phone or start with a hardware or like how deep down that particular rabbit hole did you?
Janet Vertesi:
Oh no, initially like get raspberry pies and multiple components and solder things at the kitchen table, that kind of building a phone. Because I used to build computers in the 90s. We all built computers in the 90s, right? What do you need? You need a motherboard, you need some input, you need a power source, whatever. You know how they work. If it’s a cell phone, it probably needs an antenna, it probably needs something for its SIM card, it needs network access, it needs a camera, and a microphone, like how hard could it be, right? And it turns out actually it’s hard. It turns out it’s hard. It’s not that difficult to do a sort of basic version, but the full functionality that you would expect from a smartphone, that takes hundreds of people to build. It’s like nobody builds a Mars Rover by themselves. It’s actually a very large team.
So that’s when I started hacking phones and started putting alternative operating systems on and trying to support what else was out there. And yeah.
Ethan Zuckerman:
So what sort of phone do you carry now? And how do you render it so it’s not a surveillance device?
Janet Vertesi:
Oh Ethan, I’m so glad you asked. I have a Sailfish phone. Sailfish is the operating system made by the people formerly known as Nokia. They built the latest of the smartphone operating systems to come to market in 2011. And then we’re promptly bought by Microsoft who shelved the whole project. So the guys who were building it took the intellectual property, started their own company. And that’s the company’s called Yola and their product is called Sailfish.
What’s great about it is it is a Linux phone. It’s built on Linux. The idea is you should be able to get into the guts of your phone if you want or not if you don’t. It’s also absolutely beautiful and kind of a dream to operate. And I don’t actually really have to worry so much about the surveillance because it’s already built out from the beginning. My phone doesn’t listen to me. I don’t get creepy ads. I don’t get strange connections happening because my phone was listening into something. I just don’t have any of that. It’s a thing that I kind of don’t have to worry about because it’s taken care of. The operating system is already built on the same values that I have. This is my sixth, I think, Sailfish phone. I love them.
Janet Vertesi:
So one of the big things I didn’t mention is cost. And cost can be prohibitive. Cost can be in terms of time or in terms of money. And I’m constantly turning down those little discounts at the till. I’m purchasing components from Ada Fruit and throwing money away at projects that I’m hoping might succeed and so on. Now I think of myself as trying to invest in alternatives and hold space open for alternatives, but not everybody can live like this. It is a privilege. And that’s something I want to fight. But at this stage, one of the big considerations is just, can I afford this? Is this a fight I can fight? And is it worth the cost?
Ethan Zuckerman:
You said something a couple minutes back that seems to connect this experiment to your early writing on large tech teams. And so you made the comment that you can’t build your own cell phone. That actually requires teams of hundreds or thousands. And that it’s more analogous to building a Mars Rover. That’s not a project presumably that a single person had in her head. That was a whole lot of very smart engineers working on subsystems and an enormous amount of coordination work to sort of make it happen.
When you look at something like Sailfish, and when you look at this sort of community of users that you find yourself independent with, do you find yourself sort of thinking about the next sort of tack and sociology study on how communities build projects?
Janet Vertesi:
So in some ways, you know, I’ve never really studied the alternative tech ecosystem, but in other ways, I’m an organizational sociologist of technology. I’ve worked at NASA for 15 years as an ethnographer embedded with team after team after team of people that are operating spacecraft or building or imagining spacecraft that are exploring the solar system. And what that means is I have an opportunity to use my chops in a variety of different domains, remote work for one. I mean, you can’t imagine a more remote workplace than like trying to go to work on Mars every day.
But also that these are distributed collaborations all over the world. And they have their own unique social structures as well. So, the original Mars Rover team, uh, Spirit and Opportunity, that was a collective. That team operated like a collective. They didn’t do anything unless everybody agreed. Very specific tools socially for developing consensus behind the scenes, to enable everybody to move forward with agreement.
Another mission I worked on, the Cassini mission to Saturn, that was, it was less important to speak with one voice than it was to speak with many voices to hear lots of different sides to a problem and very vigorous debate. These are two just very different approaches to what governance is. You can see this in open source communities too. Some of them are like, “I want to hear what everyone has to say. It’s going to take us three years, but we’ll figure this out.” Others are like, “Okay, we’re going to go our separate ways. You’ve got your own fork over there and we’re going to do this thing over here.” In somewhere like, look, I’m in charge of this project. I’ve decided X.
And we see this also, not just in community software, we see this also in the industrial world as well. And we saw this at NASA as well. So I think what was so interesting for me, thinking about data sovereignty, was working with these teams that organized so differently. And there were a couple of insights. One was, of course, you can’t do it alone. You have to work with people and when you work with people you form an organization But you also form relationships and that’s probably the second thing. These robots are very complex creatures, but they’re not stealing our data like okay? Also they’re not there to exploit us They’re exploring with us and they’re also You know they’re they have very limited artificial intelligence No one was like, “Let’s write Chat GPT for the Mars Rover, so it can tell us about Mars.”
No, they know that when you put a human machine partnership together, the key thing is the partnership. What do machines do well? What do people do well? What do groups do well? What do we need a group for? And then how do we make that work together in some kind of synergy? And that’s what was so cool about working with those teams is they just, they each of them had a unique way of putting those puzzle pieces together, but in every single one the technology was a thing bringing them together, not pulling them apart.
And I look at that and just look at what’s happening online today and think, my goodness, couldn’t we learn a lesson here from NASA?
Ethan Zuckerman:
It does feel like business models have something to do with all of this. And certainly surveillance has been really driven in the United States for the last 10, 15 years or so by what Zuboff has termed surveillance capitalism. And I think a lot of us have picked up the term whether or not we fully agree with Zuboff’s analysis. It feels like there’s another thing going on right now with the overturning of Roe v. Wade and the possibility of surveillance around reproductive health, not to draw you back to the story around your first pregnancy, but how are you thinking about the opt-out experiment at a moment where it feels like we’re having our first major conversation about privacy online since Snowden around this question of women being arrested for control of their reproductive selves.
Janet Vertesi:
It’s a great question. You know, when I first did the pregnancy experiment—and I should mention I never stopped once I stopped being pregnant. You know, I just had this baby and I thought instead it wasn’t about just me getting ads, it was about the baby getting ads and it was about what kind of life is this child going to lead and how can I protect them from this, right? It’s now just a way of life for me with my family. So I’m happy to talk more about that.
But when I first was pregnant, people would be like, “But why do you care so much?” You know, what’s wrong with being pregnant? Like you’re married, you know, you’re socioeconomically, like not in any difficulty, like what’s so bad about pregnancy that you feel like you have to keep it secret? And I would say it isn’t about pregnancy per se. It’s about the right to keep a secret.
And it’s also about the fact that regime’s change. You know, I have family that lived perfectly happily in Hungary that was totally integrated until the Nazis rolled in and then half of them were exterminated because they were Jewish. It was no crime to be Jewish before then, but all of a sudden it was. And you never know which civil liberties are going to be taken away.
I remember being on a panel about this at CSCW, Computer Support Cooperative Work, it was probably around 2013 or 2014 or so. And I was totally the negative Nancy on that panel, right? Everyone was talking about. “It’s so great. People are, look at what’s happening on these gay dating sites like Grindr and they’re sharing their HIV status and it’s nice that there’s an affordance for that and the technology and it’s so great moms are sharing all this data about their kids on Facebook ’cause they’re finding the help they need.”
And I’m going, you assume that because you have these rights now you will always have those rights. And one of the big things the opt out experiment is about is not everybody enjoys that right, first of all. nd secondly, those that do may have it taken away. When you’re in a subordinate position in a society, you are always vulnerable. And so it was so interesting for me about the overturning of Roe versus Wade, has it finally happened for pregnancy? Right, I hadn’t actually expected it to happen for pregnancy, but it did. And it was the first time I think people had to really take seriously, “Oh yeah, the data you gave away for this purpose has now become that purpose.”
And of course, that’s contextual integrity. Like Helen Nissenbaum’s been talking about that for 15 years. But it was the first time people saw that on their own period trackers. They were like, “Oh, I am giving myself away.”
Ethan Zuckerman:
So, but just to finish the point, you were, your principled stance on all of this has been that you never know who’s going to be the minoritized voice who needs protection. And, you know, arguably this could have been an issue for gay and lesbian people, as you have social networks revealed online, I’ve certainly done work in parts of the world where sexual identity status is incredibly important information you have to keep secret. As you pointed out, under Roe v. Wade, we’ve now given to the point where the fact that you were ovulating might turn out to be evidence in the case of ending a pregnancy.
So is this finally gonna bring about the reckoning with computers and privacy that many of us have been hoping for for years in the wake of Snowden?
Janet Vertesi:
Actually, I am not sure that the Roe vs. Wade decision will. I think it will go into the pile of things that we suddenly are worried about, but don’t know what to do about. Which is partly why I started writing about my opt out experiments on optoutproject.net because I wanted to share information with people about how to do this. In particular, I wrote a whole thing about a whole series of posts about how to keep a pregnancy private. The site is not search engine optimized, so it has to kind of be shared peer-to-peer invisibly, but it does not track because the hope is that there will be people out there who need that information and they should be able to get it.
But actually, the thing I think that’s going to bring about the privacy change is Elon Musk’s acquisition of Twitter. Right, that was the moment where people suddenly went, “Oh, this platform that I thought was for X is now actually gonna be for Y.” And I don’t know what that Y is, but I don’t like it and I’m not sticking around. And that’s been the first massive push of people towards actually decentralized tools like the Fediverse.
Now, if you’ve been an opt-out advocate for years, like me, you’re cheering on the sidelines as I’m sure you are, Ethan, to see so many people discover this and discover, “Oh, wait a minute, I can’t just assume this works like Twitter, just ’cause it looks like it. I have to donate time or energy or money to running the servers, to dealing with problematic participants in our midst. I have to suddenly, I have to grapple with racism online as a user. I can’t just assume some contracted workforce out there is gonna take care of that for me.”
And watching more and more people, not just grapple with that but come to appreciate it, I think is going to do more for the change that we want to see in imagining a different internet, an internet that we had imagined for many years that was taken over by these surveillance capitalist interests and it didn’t have to be that way.
Ethan Zuckerman:
Well, Jennifer Vertesi, that’s a great provocation to land on at the end of this. Such a pleasure or having you on Reimagining the Internet. Thank you so much for being with us.
Janet Vertesi:
Thanks so much Ethan, it’s a pleasure.