Increasingly, our daily lives are online, especially during the coronavirus shutdown. But access to the internet is not always equal, which was exacerbated by the shutdown. And the privacy implications of our online lives are sometimes an afterthought.
WPSU talked with Sascha Meinrath, who studies telecommunications policy, cybersecurity and broadband connectivity. He’s the Palmer Chair in Telecommunications at Penn State and the director of X-Lab, a think tank focusing on the intersection of technology and public policy. He is also a co-founder of the Measurement Lab, an open source project which collects and provides internet performance data for consumers and researchers.
TRANSCRIPT:
Min Xian: Welcome to Take Note on WPSU. I’m Min Xian.
Increasingly, our daily lives are online, especially during the coronavirus shutdown. But access to the internet is not always equal. And the privacy implications of our online lives are sometimes an afterthought.
Sascha Meinrath studies telecommunications policy, cybersecurity and broadband connectivity. He’s the Palmer Chair in Telecommunications at Penn State and the director of X-Lab, a think tank focusing on the intersection of technology and public policy. He is also a co-founder of the Measurement Lab, an open source project which collects and provides internet performance data for consumers and researchers.
Sascha Meinrath, welcome to Take Note.
Sascha Meinrath: Thank you. It's great to be here.
Min Xian: We're in the midst of COVID-19 and there has been increasing interest in what information technology can do to help mitigate the infectious disease. Using cellular data or even facial recognition to do contact tracing is an idea that's getting a lot of attention and already some investment. How much do we know so far about how effective these methods could be?
Sascha Meinrath: Yeah, so the ideal here is to use technology to trace, in essence, the pathways that an infection or a pandemic might take, and then to, in essence, intervene across that pathway and thus lessen the public health impacts. Thatideal, however, is not born out by the reality, which is that we don't have good data that any of this is very useful. And on the other hand, we do know for fact that this is incredibly privacy invasive. Tracking your every move, tracking everyone you come into contact with is almost by definition an invasion of privacy.
So what we're being asked to do is to make a trade off, to say, okay, look, you're going to give up these privacy, these private areas of your life in order to have this public good, this increased public health outcome, but we don't actually have any information showing that that outcome is real. And therefore, that trade off is worth making. And so that's where we are right now today is we're being sold, in essence, a result that is not empirically based, and it's certainly not born out and is therefore not worth the trade off.
Min Xian: I'm hearing two points here, which are one, this method is far from effective. And two, it could open up potential for abuse. So far, some countries like South Korea have tried using technology for contact tracing. What have we learned from those experiments?
Sascha Meinrath: Yeah. So South Korea and others have put this into place and they have found that it is not working, that in essence, not enough people have opted in, that you're getting a lot of false positives. So you can imagine in your everyday life, if you're walking down the street and you're using a mechanism like Bluetooth, that in essence can go yeah, a few feet, but sometimes it can go 50 feet or a hundred feet. And therefore everyone that you pass within 50 or a hundred feet is a contact. Well, you would get a lot of false positives. A lot of times where people would then be told, Oh, like, you've come into contact with somebody who's infected. When, in fact, they may have been a block away from you. That is not helpful diagnostic information. So that has not worked.
Here in the United States, we've already had a couple of states, uh, some of whom have paid apparently millions of dollars to utilize some of these contact tracing systems. And what they have found is that, and what they expect is this only becomes a useful mechanism once you have a critical mass of buy-in, once you have enough saturation of the population. And what they've been saying is around 60% of people need to opt in. And what they're finding is it's about one to two percent that have done. So in that regard - so you then end up with all these huge holes in your data that make it not very useful. I have a lot of questions. I agree with the ideal of the outcome that we should use technology to enhance public health. I just don't see this being actually efficacious towards that outcome. And instead, what I see it as is incredibly privacy invasive in ways that have profound implications, that these data once collected are almost assuredly going to then be abused.
Min Xian: The coronavirus exposed a lot of existing issues in our world. One of them is the lack of accessible and reliable broadband internet across the country. I understand you and your lab have found that the U S received slower internet speeds overall, during this shutdown. Can you tell us more?
Sascha Meinrath: Yeah. So, you know, broadband, cell phone, technology, all of these sort of upsides - this liberatory and empowering potential of technology has a very dark side as well, which is that if you're not on the receiving end of all of these benefits, then in essence, you're sort of doubly disadvantaged. You know, I can hypothesize about why this problem has grown so substantially, but what our data looking at tens of millions of broadband speed tests from across the United States from right before February of 2020 versus right after the stay at home orders. So in this case, February versus April, 2020, what we see is in many locations in the United States, substantial slowdowns of median speeds and what this means is that on average, if you will, the, the 50th percentile of Americans living in many locations in the U S are seeing major hits to the speeds that they actually receive in their homes.
Now why that's happening? Well, I think it's definitely because a lot of people are doing telepresence activities and that's putting strain on the residential networks. That in fact, when you look at say office connections versus home connections, it turns out most offices have better connections than your house. And so when you shift, even if we were doing the same kinds of activities at home versus our office, that would put a strain on the network, but the fact that we're doing so much telepresence, which requires a lot of bandwidth, right? Video - live video - requires a lot of bandwidth. Uh, that's putting even more strain and that's having a major impact across the U.S. to the point that we're seeing slow down speeds across the U.S. of around five to ten percent, which means for every person that's not receiving any slow down in their speeds, there's somebody that's receiving a very large slowdown to their speeds.
And we're in the midst now of investigating who these constituents are, that are receiving the worst service post-COVID. And what I can tell you in our preliminary analysis is that it's specific states, it's specific locales. It potentially is even specific internet service providers and certain service provision technologies like wireless networks versus fiber networks. For example, wireless being much more detrimentally impacted than a fiber plant.
Min Xian: Is it fair to ask: Shouldn’t we have the technical capacity for all of these increased internet usage in this day and age?
Sascha Meinrath: Yeah. So if you look historically, the United States was kind of number one on the planet for the first several decades of the internet development. And in fact, straight through like the early 2000s, really, we were number one. And we could say that and be mostly correct in saying like, we are the fastest place, we've got the lowest cost, we've got the most ubiquitous connectivity, like a lot of great metrics we were near or the top of the class. And then starting around 2005, we kind of failed. We had a massive failure. We went from number one to sort of middling amongst highly industrialized nations. That is a remarkable feat of ineptitude, over a very short period of time.
And what we've seen is that in essence, other countries, and now a growing list of other countries that are doing far better than the United States, kind of use one of two big levers that they pull: So they either ensure that there is a highly competitive market that you have many internet service providers to choose from, or they ensure that you have substantial government oversight and corporate accountability to those overseers to provide low cost, high speed, ubiquitous connectivity. It's one or the other, either a competitive market space or government oversight. What the U.S. has done on the other hand is to, in essence, pretend that we have a highly competitive market space and therefore don't need government oversight. The end result being, we pay more for worse service in fewer locations than our growing list of other countries around the globe.
That sounds absurd on the face of it. Why would we pay more for less? But it's sort of this American exceptionalism that cuts across a number of different sectors. Look at healthcare, for example, where we're paying the most on the planet per capita and have some of the worst of at least the highly industrialized nations' health outcomes.
Min Xian: The digital divide has long existed and didn't happen just because of the coronavirus. There are a lot of underlying problems, as you mentioned, including lack of meaningful competition and lack of government oversight. What are some other factors that contribute to this situation as well?
Sascha Meinrath: So there's a lot of different factors that are at play here and it's everything from difficult topology, right? We have a vast country with lots of open spaces, to a hodgepodge of different state rules and regulations, and often local rules and, you know, permitting issues and all of that - that is both real, but also rather modest, in terms of accounting for things. And I can say that because, historically, we've always overcome these problems, right? When you look at the telephone network, which is near universal, when you look at the electrical grid, which is near universal, when you'll get roads or our education systems, these are all systems that face the same issues as telecommunications, in terms of the complexity of the American topology and political sphere.
Really, what it comes down to is America just hasn't prioritized this particular infrastructure as mission critical, as mission critical to people, no matter where they're living; as mission critical to the health and growth of a thriving economy across the entire United States; as mission critical to education; as mission critical, to in essence, our social and psychological wellbeing. And as such, we've really deprioritized investment in this infrastructure and we've deprioritized oversight and accountability over this infrastructure.
And so, whereas there's lots of reasons that you could point to, to say, maybe this has an impact, or that has an impact, and they all do, but that's all marginal. That's all marginal compared to the need for bold, straightforward, no nonsense leadership in pushing an agenda that says, “Hey, universal, affordable connectivity for everyone everywhere has to be our national priority. And now we're going to take it seriously.”
Min Xian: There's a lack of transparency in broadband pricing data in the U.S. and I understand you're working on a study to map that information. Can you tell us more about the study? What do you hope to achieve?
Sascha Meinrath: Sure. So we've been collecting broadband speed data now for over a decade via the measurementlab.net platform. If you go to Google and type in, you know, broadband speed test into your search bar, a box will appear above your search results, that says, you know, do you want to run a speed test? If you click that button, it will run a speed test. And that data will end up in our public repositories, which are used by the scientific and research community to kind of figure out what is actually happening on the internet today. Now that process of collecting a multitude - millions, tens of millions, and then hundreds of millions - of broadband tests, is incredibly important. It's necessary, but insufficient to really understanding what's happening when it comes to broadband connectivity in the U.S.
It's very clear - we know this anecdotally, for certainty, that there are widely discrepant pricing regimes when it comes to broadband connectivity. We know for sure that people that might be living right next door to each other have wildly different bills for exactly the same service. We know that there is rent seeking amongst a lot of the different realms of broadband service provision, transit, peering, all these other elements that intertwined this fabric that is the internet. And we know that we have almost no transparency of any of this, that as a consumer living almost anywhere in the United States today, if you want to do an apples to apples comparison about what broadband services are available to me and what are their costs, it's like, you should get a PhD if you can figure that out, because it's so darn difficult, needlessly complicated. And I equate this to, imagine, if you have to like go buy your groceries, but you had no idea what anything costs. And they would just tell you, once you got to the checkout line, what your bill is, and you may or may not be charged the same as the person, right behind you for exactly the same goods. We would look at that and say, that is an absurd system, but that is what we have with broadband connectivity, that the information is held by the internet service providers and is entirely opaque to us, the broadband consumer.
So how can we then make an informed decision? Where is broadband? How much does it cost? Without those two pieces of information, I don't know how you make informed decisions about, for example, how to bridge the digital divide, because you don't even know where it is or what's causing it, or the extent. And that's where we are today, that we have, in essence, we all fully self blinded ourselves and in so doing, have enabled all sorts of business models and practices that have not been in the best interest of consumers. And I look at all of that taken as a whole, and I say, well, this is probably a sub optimal outcome that we should probably address forthrightly and immediately. And the coronavirus just adds kind of fuel to that fire. It adds really kind of a catalyzing agent to say, Holy cow, we have to address this problem because now, the detrimental impacts are so plain to see that even without the systematic, empirical evidentiary trail, we can't ignore this any longer.
Min Xian: If you’re just joining us, we’re talking with Sascha Meinrath, the Palmer Chair in Telecommunications at Penn State. He’s the director of X-Lab and a co-founder of the Measurement Lab, which collects and provides internet performance data for consumers and researchers. His research focuses on telecommunications policy, cybersecurity and broadband connectivity.
You have done extensive work on telecommunications policies, including privacy and cybersecurity. On May 14th, the Senate voted to reauthorize the USA Freedom Act, which restores our government's surveillance powers, and would allow collection of internet search and web browsing data without a warrant. Can you talk about what this means to everyday Americans? Why is this significant or some would say problematic?
Sascha Meinrath: Yeah. So most of us assume that if we're sitting in our house and doing something on our home computer, it's private, that surely there's no way that law enforcement would get access to that information. And surely they would not be allowed to get access to that information without getting a warrant. But the reality is that the U.S. government has declared that search histories, which we look for online, they can grab without a warrant and they can scoop that up in mass. Now, there are many of us that say, look, that is clearly a search of my private possessions and private information. And at the very least before even get into whether you should be doing this in the first place, at the very least, you should have to get a warrant for collecting this kind of private information about what I'm doing inside my house.
And weirdly we are currently amidst a policy battle over whether that is in fact the case. So this is what we know of, this Section 215 of the Patriot Act, which has now been interpreted to mean that law enforcement can collect hundreds of millions of records on U.S. citizens who are completely innocent of doing anything under the guise of searching for terrorists, but now often used for a variety of other law enforcement activities. Doing this without any warrant is problematic. So what's happening right now, we've got a bill that would reauthorize this section of the Patriot Act for data collection, in essence, on the backend. So this would be, if you use, you know, Apple products, Google products, Microsoft products, Facebook products, all of these big mega technology company products. The question is, should they then be allowed to share that information or even sell that information to the NSA, to the National Security Agency, to the FBI, to other law enforcement agencies?
And I would argue, no. I would argue that that private information is private first and foremost, and that the Fourth Amendment of the Constitution protects against unlawful searches and seizures, which declares explicitly, and without any caveats, that you must get a warrant to execute a search on your private lives. But we are now battling over how, in essence, the rules of the road should look when it comes to reauthorizing this.
So what are the amendments to this bill that did pass, is in essence saying, look, when you go to a secret court, like the Foreign Intelligence Surveillance Act court, the FISA court, and under the Foreign Intelligence Surveillance Act court, you collect domestic intelligence that if you're the FBI and FBI that has by the Attorney General of the United States, or sorry, the Inspector General of the United States been found to have lied to this court, then it's probably good to have a defendant in that courtroom as well. That it shouldn't just be the FBI presenting to a judge saying, we need this information in a secret court, that there should be some advocate standing up for the rights of individuals on the other side of that ruling. Somehow that has become a fairly radical position in D.C.
So Section 215 was used because law enforcement said, without this in essence, terrorists will run rampant and et cetera. And so there's been a series of hearings, where the FBI and others have said, have been asked under oath, how many terrorists did this catch and what we're finding time and time and time again is in fact, these are not very efficacious programs at all. Then again, the trade-off, just like contact tracing, that we're told, you're going to give up this privacy - in this case, your web browsing histories and et cetera - and by doing so, you will enable us to catch terrorists. It's far as we can tell from the public evidence is that these are highly ineffective programs that, that trade off that were promised you give up privacy, you get safety. In fact, we don't get the safety. We've just given up our privacy.
And you know, my argument again has always been like, we should follow the data. So if we have a program of this sort, we should mandate that we collected information about the false positives. How many people did we wrongfully label as terrorists? And we should be looking at the false negatives. How many terrorists did we miss with this program? These are the kinds of questions that I feel like again, applying the scientific method to evaluating how effective a surveillance program is, how effective a public health initiative is. It's just smart. It's what we should be doing. And yet when it comes to the application of these new fangled digital technologies, we just throw out things like science, like empirically backed decision making, like informed decision making. And I think we need to bring that back to the forefront of these debates.
Min Xian: What do you say to people who understand these problems and want to make the situation better, but feel there's no way to opt out of a digital life?
Sascha Meinrath: Yeah, this is the challenge of our era. But the idea that this is more entrenched or more difficult to solve than - let's say, the civil rights movement, women enfranchisement, the end of slavery, the elimination of our British overlords, right? Like America is steeped in, born of a revolutionary zeal for freedom, for, in essence, our individual civil liberties. This is just the digital era battle that has to be fought.
Now I'm an optimist. I think we fight and we win. It doesn't mean it's going to be easy, in fact, far from it. But I actually think, looking historically, this is an easier battle to win than many of the battles that we have fought and successfully at least move the ball forward on. And so, you know, what do I tell people that feel like there's a lot here and it's a very difficult battle and they feel a little hopeless or helpless to affect meaningful change?
I would say, I hear you. It is true. This is a battle for our times, and it's going to be a long slog and it's going to be demoralizing and it's going to be difficult, but it's a surmountable problem. And it's one that as more and more individuals start talking about, “Hey, this happened to you? This happened to me.” “It has been - I'm not comfortable with this process, that program, this invasion any more than you are.” As we have these conversations more publicly, it builds pressure. And by doing that, maybe not this month, this year, but by doing that over the course of the next generation, we can affect, and we will affect major societal changes. We will create an essence, a new normative framework of what's acceptable and not acceptable within this digital era. And that's what I, and many, many other folks working in D.C., that's what we're fighting for, that we truly and honestly believe this is a winnable battle. And then it will take a lot of blood, sweat, and tears, frankly. But that we get there - the more that we're able to talk about this and to normalize the questioning of these practices and programs that have, in essence, a PR promise that's never delivered and a trade off that has been so detrimental to so many.
Min Xian: Sascha Meinrath, thank you for joining us on Take Note.
Sascha Meinrath: Been my pleasure. Thank you.
Min Xian: Sascha Meinrath is the Palmer Chair in Telecommunications at Penn State and the director of X-Lab, a think tank focusing on the intersection of technology and public policy. He is also a co-founder of the Measurement Lab, an open source project which collects and provides internet performance data for consumers and researchers. His research focuses on telecommunications policy, cybersecurity and broadband connectivity.
You can listen to more Take Note interviews on wpsu.org/takenote. I’m Min Xian, WPSU.
