librarypunk - 073 - Data Cartels feat. Sarah Lamdan

Episode Date: November 13, 2022

We’ve got Sarah Lamdan on to talk about her new book Data Cartels: The Companies That Control and Monopolize Our Information! We talk about government contracting to third party data companies, legal loopholes, the connection to academic publishing and metrics, and the stranglehold the corporate duopoly has over access to US law by lawyers, citizens, and prisoners. Justin also advances his theory that academic journals are just podcasts.    https://twitter.com/greenarchives1  Data Cartels: The Companies That Control and Monopolize Our Information | Sarah Lamdan Librarianship at the Crossroads of ICE Surveillance S3 Ep2 – Knowledge Equity Lab Media mentioned Fact Sheet: New Records Provide Details on ICE’s Mass Use of LexisNexis Accurint to Surveil Immigrants - Community Resource Hub Weapons of Math Destruction by Cathy O'Neil: 9780553418835 | PenguinRandomHouse.com: Books  Algorithms of Oppression https://twitter.com/JailLawSpeak https://libraryjuiceacademy.com/shop/course/317-zotero-for-librarians/?attribute_pa_session=2023-03-mar

I just play the theme music just to get us going. So I don't have an opening. Welcome to Library Punk. Let's go. I'm Justin. I'm a Skalkan. My pronouns are he and him. I'm Sadie. I work IT at a public library. My pronouns are they then. I'm Jay. I'm a music library director.

And my pronouns are he, him. And we have a guest. Would you like to introduce yourself? Yeah. So I'm Sarah. My pronouns are she, her. And I'm a law librarian, but also a professor of law at City University, New York School of Law. Hey, welcome. You have a new book out called Data Cartels, which Spark is trying to get in on our action. I just got an email from them saying that they're going to have a webinar that you're going to be leading. And I was like, the wreckers.

The wreckers are getting in on our turf. Yeah, I am an FOS friend of Spark. Yeah, I do a lot of work with them. especially around like antitrust stuff having to do with Elsevier, which shouldn't be a surprise. Yeah. We had Scarlet on not that long ago. Talk about contracts. Yeah.

Scarlet is great. Spark is great. I just, yeah, librarians. Spark is like the best thing that's ever happened. Oh. I love Spark. Yeah, you're right. Spark is the best thing ever happened.

It's a shame you can only become a member institutionally because I think a lot of people would enjoy being individual members. I definitely wanted to be when I was. not at a big enough library. Yeah. And we weren't Spark members when I started here. And I was like, oh my God,

to get the paltry amount of money, it's like $6,000. And I was like, okay, but what if we split it among the three divisions and we do these complicated budget transfers

for $2,000 per division? And like, I'm the only librarian at my institution. Yeah. So it's like, you know. That's a bummer.

Small institutions, yeah. Yeah. Okay. So we're going to get into data cartels, but I have a quick Reddit. Ask Reddit. I don't know what I've called this. Reddit. Those people are dumb-dums.

Okay. The post is Library Tech. Now what? I'm a library tech in Canada that works for a university. As far as LibTechs go, at this university, I'm currently at the highest level. So now I'm not sure what to do, and I haven't been working at this university for very long. The place I live

is extremely expensive, so it would be different if my current job paid enough, but it covers enough for bills, but not really savings extras. So I'm wondering, what have other library technicians done? Are there other library fields, public, private law, et cetera, that maybe have more growth opportunities than academic libraries? Have people gone on to do their MLAS after being a library tech?

I'm feeling kind of unsure stuck, so I just wanted to hear some others' opinions of what their situation is like. I think we've talked about this, particularly when we talked about party girl. and the pressure to go to library school. Yeah, a lot of the people that I know who became librarians since I've been working in libraries, started working at the library first as a tech or an assistant or something along those lines. Yeah, I dodged that bullet by going, what if I just go into library IT instead? But, yeah, getting your MILS. It's a lot of school.

it's a lot of school and that's why I didn't go down that path so I mean maybe with like the growth thing I mean if this person's willing to go into like a corporate sector there's a lot of like companies that need like taxonomists and whatnot you know a lot of the skills we learn as librarians are actually very valuable to like like I was going to take a taxonomy course in grad school and the person who teaches that was the taxonomist at Etsy stuff like that like maybe don't go work for like war crimes ink but you know like those have more growth I've seen several especially like tech services librarians lately leaving the field and going into either like software development or working in techie but you know data related positions where it's using the skills they got in their library science degrees but you take the

meta off the front of metadata and all of a sudden you're making over 100,000 a year. So I know that that's not always the kind of desired environment people want to go into, but if you're looking for like the library tech stuff and I don't know if getting the MLIS would then be necessary. Like you could maybe get another type of degree if you're interested in these kinds of skills. But I also do not have this experience. So this is just me going off of like what I've seen my friends go through. So. It seems like they already have a BA because they're just saying do the MLIS. And I think some places to be a library tech, you have to have a BA.

Like, that's the situation for our library assistance. But, like, you can get, like, two years, master's or certifications in, like, various types of, like, data science and comps. And stuff like that that don't require bachelor's degrees in those things. And there's also plenty of, like, free online resources to get you copy. up if you're not quite there yet. And they're kind of fun to go through. I've gone through some of them. Yeah, I would say use your job to get as many skills as possible before you leave and then see if you can turn that into like a digital asset management job. Yeah. Those are pretty, even with the tech downturn, I would say there's still a need for people to do DAM because corporate librarianship

died like 20 years ago. Yeah. And so now they're hiring a bunch of DAM special. who I have just learned are informed that they have just learned metadata as a thing, as a field. So good for them. Can I give them my number and make money off of like consulting for them or something? Yeah. Go present at a conference about digital asset management and see if you get hired. Yeah.

That's the consulting racket. Yeah. I say racket. I'm trying to do it too. Yeah. I mean, my library juice academy course is posted now. So, you know. When does it start?

March. Yeah. I can be gross and plug it at the end if you want. Sure. Yeah, it's gross. Yeah. You hear that Sarah, plugs are gross. Don't you fucking dare plug your book?

Fucking sell out. I have nothing to advertise at all. Okay. Well, that was Reddit Ask Reddit. So, Sarah, we've brought you on to talk about your book that you definitely can't plug called Data Cartels. which is pretty relevant. It's also the subtitle is the companies that control and monopolize our information. And you've been doing this work for a while.

I've vaguely been aware of it since my job is scholarly communications. And this is just the kind of world I live in. When I got to the chapter on academic libraries, I was looking to the footnotes for some reason. I was just like, I know them, I know them, I know them, I know them. I know them. So. Yeah. A lot of schoolcom happening there. Yeah.

Yeah, so I actually, yeah, I started the book. I didn't start like, oh, I'm going to write a book, but I started thinking about this stuff because I was a librarian. So notice I'm not a librarian anymore. That's in part because of the book. I don't know if you want me to tell the quote unquote story, but I can explain that if you want. Yeah, it's the first question. How did you come to this work? So it was 2017 and I was a law librarian at City University in New York. And one of my main jobs was to teach people, everyone how to use Westlaw and Lexus. And I was, it was 20, it was like right after Trump was elected and ICE was in the news a lot because it was doing just heinous, egregious things as an agency. And Donald Trump had passed an executive order saying that he wanted to do. extreme vetting, which I guess ICE took to mean intensive data surveillance. And so ICE had an investor day. And it's kind of a shot across about a few reporters FOIA, the list of companies that wanted to, you know, be a part of this extreme betting program. 2017, ICE bad, ICE terrible, companies trying to do extreme betting. And the FOIA list had Lexus Nexus reps on it. And it also had Thompson Reuters reps. And Westlaw is owned by Thompson Reuters and Lexus is owned by Redall Severe Lexus Nexus

or Lexus Nexus Nexus Nexus Nexus, Lexus Nexas, vying for ICE contracts. And at that point, I didn't know anything about Lexis Nexas and Westlaw except for that they were legal information platforms. They're the only two legal information platforms the lawyers use. And I was like, wait, what the heck are they doing with ICE? And so a colleague and I wrote a blog post for the American an association of law libraries. And we didn't, like, it wasn't, I mean, really, in, in response to everything that was happening, it was a really, really tame post. It was basically like, hey, we noticed this. This seems bad. Do like, should we care about this? And AALL, like a law library posted it on the blog. And then within two minutes, AALL took it down. They like censored it. They're like,

no, you are not allowed to talk about that in our professional organization. And I was like, are you like are you censoring me that i pay thousands of dollars to be in this organization and i am a member like this is our professional organization where we communicate all of our professional you know questions and learning and um they were like nope you can't you can't criticize you can't criticize these companies that way and that pissed me off and so i started researching because i really wanted to figure out what was going on and i wrote like i wrote just an academic article because i learned that thompson readers and read else of your lexasus nexus is don't just provide the only two major legal information platforms and legal research platforms. They're also two of the government's biggest data brokers that work.

They build, basically, they build predictive policing products and other policing data analytics products. And they also have these huge troves of personal data. Like they have data dossiers about all of us. And they sell those to tons of law enforcement agencies. So I kind of discussed that in a law review article. And at that point, so every law school has like a LexisN. a group of LexisNexis representatives and Westlaw representatives, like they pay people to kind of bring their brand to our school. And Lexus Nexus switched out the rep that we'd had for

like 20 years and brought in somebody new. And the new person basically, it seemed like was brought in just to kind of watch me. And they started demanding, they started calling my boss, my library boss, the dean and demanding to meet with me. They told my students that I was lying. It was just wild. And then eventually, I mean, there were a number, I was, there were a number of reasons that I changed my job. But one of them was that it became really difficult for me to work in the library because it was really disruptive. So they just moved me onto a professor faculty line. And now I'm not a librarian at my job anymore.

So yeah, that's that is how. And all of that made me really, I mean, you know, it made me angry, but it also made me curious. and I've been researching since 2017 to write this book. It was your institution supportive of you, and that's why they made the move? They were supportive. And I've been teaching some classes. Like, I've been kind of doing, you know, there were, it wasn't just, this wasn't the sole reason that I was moved out of the library. And they were supportive of me, but the most important thing, and I say this to all librarians,

the most important thing about my position is I was on a tenure line and I had gotten tenure. So there was no way anybody could get rid of me, right? I'm a tenure law. I'm a tenure law librarian. Tenure. Tenure and unions. But yeah, tenure is pretty much what made it so that, I mean, they were supportive also. I work at a pretty cool law school.

And everybody was supportive of, you know, my work. But also, there wasn't that much they could do. When the Nexus rep was talking to your students, I mean, is it because you were teaching a lot of classes and you were bringing, because you were teaching how to use? Lexus and that was why they were contacting your students. So like, because as far as I don't remember you publishing all that much about this. I didn't. And in fact, I was very careful about what I published because initially it felt really intimidating to write about this because so there are maybe, I think at the time,

maybe like 2,000 librarians in AAL and all of them became afraid to talk about because nobody, I mean, it's very rare to have tenure as a law librarian. And most of my friends didn't have that kind of, you know, most of my colleagues didn't have that kind of protection. So librarians became really afraid to talk about it at all. And so it became kind of like just in our community. It was like intimidating. Nobody, nobody wanted to talk about it. Nobody could talk about it. So I didn't want to talk about it too much because I didn't want to, I didn't want to screw other librarians over. I didn't want to harm anyone. So I didn't talk about it in library world very much. And then I have to say, like, immigration lawyers were interested,

but a lot of lawyers didn't seem very interested in the problem because it seemed kind of niche. It seemed kind of unique to research into librarians. Honestly, it wasn't until I started working with Spark. Spark cared. And that really, for me, changed a lot and made me feel more comfortable doing more about it. But yeah, I was really quiet about it. I mean, if you're at a job and then all the sudden, some people are like demanding that you have meetings with them and talking to your boss. It just, it makes you kind of say, okay, you know what, forget it. I'm not going to say, I'm just whatever. I'm not going to mess with it. You get kind of, it's not that I wasn't afraid to talk, but I just didn't want to mess with like all the drama associated with it.

Yeah. I would see myself just ignoring. I would, I would think my boss would probably just also hang up on someone doing that to me. If I was just like, I'm, this person's just harassing me. Just ignore them. Yeah. We would just shut them out. But yeah, I mean, you only get like one rep. So it's really annoying. Yeah. Yeah. And I didn't want that rep to keep bothering everyone. And also, I, you know, it was fine for me to move, you know, move to the faculty to that was, there were other, you know, like, luckily, there were extenuating circumstances and it was a comfortable transition.

But I think other librarians who saw that happen, who didn't have the protection of tenure, then realized that they didn't want to say anything at all, which it makes sense, right? If you're not in my position. and then some brand manager starts calling your boss all the time. I get it. Yeah. So I read through the book today, and a lot of it is pretty familiar if you're in this world. But I think the big takeaway, at least for me, is that there are these data brokers that we are aware of because they're in the news, which is like Clearview, AI, Palantir, what was it, Cambridge Analytica, like all those like bogeymen showed up. And, like, you know, Facebook gets dragged in front of, like, the Senate and Twitter gets dragged in front of the Senate.

But Relics and Reuters, Thompson Reuters, don't really get this scrutiny and are treated as if they are obscure things, even though they are much bigger than these other data analytics companies. And they control more industries. And the only reason that those industries work is because they get data from LexisNexis. and from Thompson Reuters. So I think if this could get some more popular attention, you know, maybe we could see some of them get dragged in front of Senate hearing committees and have to explain like, why do you control all of U.S. law, basically,

and case law and civil and court documents. And when your platform goes down, no one in law can do their job. and then also you sell, I guess the main reason is because they sell data to the government and they do the work of publishing for the government. I didn't see you mention once the government publishing office, which is my favorite federal office to shit on. Yeah. Because they don't have the power to do anything.

And they run the FDLP, which doesn't have the power to do anything. But like every library I've worked in has the FDLP and they treat it so seriously. I'm like, they can't do anything if we don't do anything. No. They can't. There's no enforcement provisions. No. And you can get rid.

We got rid of our FDLP actually. Like we were in, when I started working at the law school, we had an FDLP, you know, section and we collected all the FDLP materials. And we just, we were like,

at some point we're like, no, thank you. We're out. We're not going to do this anymore. One of the things I did as a graduate assistant when I was in grad school was to check like the FDLP like perma links. They would give you for things.

And like 99% of them were broken. all of the time. So bad. Because I had to go fix them in lib guides and stuff. Because I was like a lib guides admin as part of my job. I worked for Lisa Hincheliff, if you know her. Yeah. Wow.

Yeah. She was like my like boss in grad school. And I had to go fix FDLP links and they were all broken all the time. That sounds really fun. Yeah. Yeah. So the government doesn't know what the fuck it's doing, by the way. No.

Its infrastructure is run by like mice in 1990s. I mean, it's questionable whether or not the GPO has the congressional authority to even do the FDLP anymore, like to run digital publishing the way it does. So yeah, my last job, I convinced us to get rid of it because we were just outsourcing the cataloguing anyway. So it's just a $3,000 bill for us. And we were like, why are we doing this? We're a tiny private university. And we have FDLP at my current job and I'm like, why are we doing this? It doesn't matter.

Right. And I feel like now so much of it is just kind of, it's born digital anyway. And it's, it's like you can find it without necessarily visiting an FDLP library. It's become kind of attenuated from its purpose. I don't know. I've mixed me like because I do like, I wish we had a more robust government document system that was like really well supported and that was digitized. Like I feel like there are other countries that have these really nice. Like you go on their kind of whatever their webpage that would like introduce you to the. this massive, well-organized, beautiful library of digital materials, but we just don't, we don't build that kind of infrastructure for- There's too many states. We just don't want to do it. Yeah. I think it's part of it. But this is kind of an old problem. So Relics and Thompson writers are the original data brokers and they were law publishers and they were responding to the fact that in the 1800s, I didn't realize these companies were so old that, like, Westlaw was that old. But they were publishing case reports simply because, the government took too long to do it.

And I remember you also mentioned, like, the current Supreme Court rulings are only digital up to, like, 2012 outside of Westlaw. Right? Did I misunderstand that? The digital ones now, now the Supreme Court does a good job of posting its opinions online, right, when they come down. But the paper ones, so you can't, back in the day, they didn't digitize. And if you wanted, then, so there was this kind of gap, unless you had access to, like,

Westlaw's case. books. And this was because this was before digital. And it's still this way. Like, if you go to a law library, they'll have a paper set of the United States Code or of the Supreme Court reporter, but they stop years before whatever the current year is. Like a lot, a significant number of years, like five, seven years. And if you want anything that happens, you know, in that period of time, you need Westlaw or Lexus or you need their publications because they publish also. versions of the code. They're called the unofficial codes and then the unofficial court reporters and they they will publish them more quickly than the government does. Yeah.

Yeah. And then some states completely seed because it takes codification and printing. I mean, it takes time and it takes a certain, it takes a certain amount of organization. It's labor and cost and a lot of states just outsource that directly to the publishers. New York does. New York doesn't have like an official. code. There's Lexus version and a West version. And that was why that case happened in Georgia, because the official law was the Lexus version and someone was pirating it and the Supreme Court ruled that, no, since this is the official law of the state, that's not piracy. It can't be copyrighted. I think that was the arguments. It can't be copyrighted. Or at least it had to have

public access. It can't be copyrighted. So there's a lot. There's a lot going on. But yeah, so the actual code cannot be copyrighted. The government edicts doctrine says that like we own law, the legislature is our body that we fund and that we elect. And whatever the products they make, the laws they make belong to all of us. And they should be accessible to all of us. So what West and Lexus do is they quote unquote annotate them, they change them just enough. They modify them. They put headnotes on them. They put, you know, they link, they hyperlink to other parts of the code or put page numbers in them. And they modify them just enough so that they can say that, you know, they have copyright protection because they're original. They're modified. And that's kind of how they claim copyright over the law. And what, so it was a,

it was a John Roberts opinion. And what he said is that it's, it's not fair. to create kind of two tiers of the law. So the public can have this like outdated, crappy, cheap, free version, right? And then people who pay for Lexus can have a fancier version of the law. And he said that shouldn't be allowed. And especially parts that were deemed as constitutional. So like the example you used was the sodomy law is still on the books. But if you had the annotated version, you would know that that has been overturned, I guess. Yeah. Yeah. Exactly. If you just looked at like a free version online or like whatever version, you know, the state might put up, it would be really hard to figure out that that law had been overturned, right? You would read and you say,

oh, wow, that's, you know, that is a crime. And then if you look at, you know, the updated, nice, fancy Lexus version, you can see that it's been overturned. Yeah, it's like those websites that used to be like, crazy laws, women in Florida can't parachute on Sundays. And it's like, that probably isn't enforceable. There's one, like, you can't carry ice cream in your pocket. I forget some state. Yeah, you can hunt whales from your car. I think the ice cream in the pocket one is here in Washington. That sounds really familiar here.

I don't go about like a neck of a giraffe or something. I don't remember what it was. That was a one of one. Hey. But where do these companies get data? Because I think people will find that pretty interesting. So each chapter in the book represents a different market. it like informational market.

So I kind of put a divider between like published information, like all the academic journals that Elsevier owns. And then like the law, which gets published, you know, it's been like, it's vetted. It's, it's information because it's structured. And then they also have this whole other racket where they collect personal data and use that to create information by running it through data analytic systems. And the creepy part about their data business is they both claim. they get data from over 10,000 sources.

And they say that some of those sources are public records. So some of the sources make sense. You can figure it out. Like voting rolls, you know, marriage license records, deeds to houses, criminal records. There are certain records that, you know, are public records, aka the government makes these records and you can get them. But then there's thousands of records that they get. And we're not sure where they get it from.

You can kind of pull together an idea of where they get it from by what they advertise, like they advertise that they can help companies track social media so you know that some of the data must come from social media. You can see that a lot of downstream companies, downstream, you know, smaller data brokers or data collectors sell their data upstream. So you can imagine that Lexus and Thompson Reuters probably both buy those, you know, those data, those datasets from downstream vendors. But there are 10,000 sources and there's no listing of those sources. So one thing, when I first started doing this work, I was working with a lot of immigration organizations. And one of the things I really wanted to find was their directory of sources.

Because for all of their other informational resources, you can get like a database directory. This is such a librarian term, but like Westl used to mail us every year, a paper database directory. And it's like a phone book, right, that says everything they have that the year, Spain, you're like, we have all Washington state law from or statutes from, you know, this year to this year. And here, you know, here's the original publisher. And so you can. see exactly what is in your product. And for Elsevier, you can do the same, right? You know what journals they carry. You know what, you know what issue they're on. And you have the information about the scope and the content that they're giving you. But in their data products,

there's no directory. And it's impossible to know what 10,000 sources they're getting their data from. So one of the efforts that people do is they try to figure it out by doing freedom of Information Act requests, they FOIA ICE or they FOIA law enforcement agencies to try to see what's in the contracts or, you know, if they get some sort of directory or list. But there's no law that says that Lexus or Thompson-Ruiter, sorry, has to provide the public with a list or there's no way to see it. Yeah, it's kind of a running theme is there's no law for tech companies, even though these are very, very old businesses. But because they have these data analytics, it's considered too complicated and too fast changing to regulate.

Yeah, they sell to law enforcement and they sell to those other data sources like we mentioned. But like what other products are they selling? Because I think there's some stuff in here about like rankings, dossiers, stuff like that. What kind of products are they selling our data into? They're selling our data basically to any market they can think of, anyone that they can figure out how to sell it to. So they sell both our raw data and these data analytics products that they call risk products where they figure out algorithms and other machine learning and AI systems. I'm putting air quotes around those terms. But they build these systems that, you know,

they shove all of our data through it to make predictions about how risky we are. And they sell those products to insurance companies, Lexus Nexus is like one of the biggest, if not the biggest seller of insurance risk products for the insurance industry. They sell products to tenant screening companies. So people who decide who gets housing and they sell it to banking systems. They sell it to, I think they say they like at Lexus in one piece of documentation says that they sell to all sorts of like SMP 500 companies. So just general companies. And one they're getting really into is selling it to health care systems. So electronic records management is a big market for data brokers. And it helps if you can build products that maybe assess

whether somebody might be at risk of being an opioid user. That's one. They say they're helping to fight the opioid war by ranking people based on whether they might become addicted to opioids. And also there was one other thing that I don't remember. I think banks, too, seven out of the ten biggest banks. Yes, the banks and also bank risk. And it's all risk. So what they say they do in the banking industry is rank us by how much of a broad risk we are. Are we likely to commit fraud?

And the same with the insurance. Are we likely to commit insurance fraud? Are we likely to, you know, commit bank, other types of financial fraud? are we likely to commit a crime? They sell one of their biggest customer bases of law enforcement and sort of, you know, intelligence. Yeah. You're saying, oh, yeah, you're saying about the opioid risk. I remember reading a article a couple years ago about some doctor's office that used it,

and they actually ended up denying somebody pain meds for a surgery because they had been a victim of, sexual assault. And like it was that quick of a like line. It was this person's been through trauma, which makes them more likely to be addicted to opioids. Therefore, for this completely unrelated like hysterectomy, we're not going to give her like these certain drugs. And it was just like, it just seems so much like a re-victimization. And the other thing with it being healthcare, healthcare is like a giant ransomware risk and is only increasing too. So, you know, you've got all this data going to healthcare and all of these predictive systems, which is then turned around and made into ransom, you know, ransomed.

So that recently happened here, well, in Washington, as well as all over the country, a common spirit health care, I think it was, went through a ransomware attack and like hundreds of hospitals got all of their systems knocked offline. So, trickle-down effect. Yeah, we've talked about Ransomware recently. And so it's... Yeah, it's always on my mind. I hadn't thought about that in medical settings. And now I'm like, oh, they're a big target. Now I'm going to have nightmare, say any things. Yeah, I think I took a quote from your book. Seventy 500 federal state and local agencies have relics products. 70% of local government, 80% of federal agencies These use relics products, 2100 police departments and 955 sheriff departments,

$93 million in data program contracts with ICE and private sector analytics products to 95% of a top 100 personal insurance company, 76% of Fortune 500 and 7 of the world's top 10 banks. So they're really all over the place. And it's really strange that it's considered such a background issue. Yeah. they do a good job of making sure that their contracts are kind of opaque. There was an article recently, an immigration law organization called Just Futures Law managed to get a lot of emails about Lexus Nexas relics's work with ICE. And it turns out that Lexus Nexus puts like a non-disclosure agreement in their

contracts that prevents ICE from saying anything about Lexis Nexus and any of its public-facing documents. So I mean, I feel like if they're doing it with ICE, it's. probably something that you can imagine they're doing with other entities that they think might, you know, be bad PR to be involved with. So I think some of the worst stuff they do is stuff that we can't even see. The next bit, though, really made sense because I read Kathy O'Neill's weapons of math destruction. And that was when I first started thinking about like, oh, if you drive through this neighborhood, your GPS picks it up on your phone and that could go to your insurer and that could signal that you might be a drunk driver if you're commuting to work at 5 in the morning and drive past a bar and stuff like that.

And so that can impact your risk score. And you give the example of someone breaking really heavily to avoid a crash, but because heavy braking is associated with crashes, that could be flagged without context as a risky driving situation. And that kind of leads me to what I want to talk about, which was the no-context aggregation of data. So you were saying it's a big issue that if you are already in the system because you are on SNAP or you are getting cash assistance or you are on any welfare program, there's more data about you in these data brokers. And that data then goes to law enforcement agency. So you're more likely to be picked up for predictive policing, child welfare. and insurance risks.

So the data doesn't have any context, but they have more data about you. So it perpetuates the inequalities. And I think you also cited algorithms of injustice, or I can't remember the title of the book. Algorithms of oppression. Yeah. Yeah.

So there's a lot of scholarship and a lot of research about how, there are two places where these systems are really, really biased, right? One of them is algorithmic bias, which is algorithms of oppression. It's Sophia Imogian Noble's work. And I know, right? Yeah, I feel the same. Can we get her on?

Can we try? That would be amazing. I'll try, yeah. Cool. Work my gift to gab. Yeah, she's awesome. And so there's that, you know, there's that kind of bias, algorithmic bias. And then there's also data bias, right?

So data is also flawed. There's no pure, perfect data set. Data is riddled with errors. And then data also is, it's not even in the sense that some people have a lot more data in particular collections than others. That's especially true in the criminal context and in context where there's a lot of government contact because people who tend to be overly policed, people of color, black people, Muslim communities, people who live in, you know, socioeconomically less wealthy areas, they are overpoliced, over surveilled. That means that

their police data dossiers files are huge compared to people, you know, who are white, who live in affluent neighborhoods, who live this kind of police free or generally police free lifestyle. And that is, you know, it's a form of systemic racism that works its way into, you know, gang databases and, you know, arrests and stops, stop and frisk, obviously in New York City was a huge campaign to stop that kind of systemic injustice. But when you take information from those stops and information from gang databases and you datify them, then you're just perpetuating those same types of systemic injustice in your data sets. And also, like you said, people who have more contact with government, so people who are entwined with child protective

services or entwined with other supports, like you said, like SNAP, food stamps, other things where the government can suddenly track a lot of what you're doing, right? That creates a record about you, too. So you take that. inherently biased data and you run it through the inherently biased algorithms and you get just kind of a double discrimination or, you know, exponential discrimination because you're combining two things that are already totally biased together to make one list of the most risky people, right? The people most likely to commit a crime. Yeah, one study that I cite in my book about this algorithmic predictive policing program, I think it was a predictive policing program that was

implemented in Chicago. And at the end of the program, really, all they found out was that they kept arresting or focusing on targeting the same people over and over again because there were a few names that were just really overrepresented in their data sets. So those people got unfairly surveilled and were always under kind of the police target. Like, if people not fucking watch Minority Report, like. Exactly. Right.

It's right there. Like, it's a good movie. That wasn't meant to be a minority report. blueprint for the future, dear God. I'll admit, I didn't watch it for years, and finally, a few months ago, my partner was like, you have to watch it. And we sat down and watched it, and I was like,

oh my gosh, it's so good. It's so good. And it's true. And now it's like, yeah, that's like, what Lexus and Nexus is the building. So it's fine. It's fine. It's fine. Let's make a palantir. That's not like a thing that

Sauron uses and Lord the Rings for evil. Now we're just going to go full Panopticon, Foucomo. Let's make Panopto and sell it to universities. And when Justin says, why is it named Panopto? And no one seems to see it that that's a problem. Whatever, fine. What's that?

There's a whole joke about like the worst tech sci-fi thing you can invent is like to you, you're like, oh, this is the worst. This is dystopia. And to some tech person, it's like, oh, this is amazing. And we could totally make bank off of it. You've just invented the I have no mouth and I can't scream box. Exactly. The thing you stick your hand in in Dune.

I also forgot to mention to keep us on our theme for the last episode, next couple episodes, is disability. So if you have disability, we talked about biocertification. And so all the extra work you have to do to interact with the government to prove you're disabled or your employer, anything. All of that is more contact points with the government. And you mentioned disability as well. I just forgot to bring it up. Yeah. I was going to say the one scene for minority part that always stuck with me was when he's

walking through like that public session and they're like, Iris scanning him and selling him ads directly. And it's announcing his name. And yeah, every time I think about biometric multifactor authentication, I think about that scene. I thought the other day, if I, you know that guy, have you seen this man in your dreams? that guy with the big bushy eyebrows. What if I got a tattoo of that on my thumb and I used it for all of my picture authentication?

So the reason these companies can do all this stuff is because they are third parties and they are not considered government entities and this is the third party doctrine which means that they can basically more or less privatize and commodify data as much as they want to without being really touched by any of these laws, even like the laws that like Equifax have to deal with.

So like the privacy law or the financial data law. And so if we were to want to apply these laws to these companies, I mean, if we like nationalize them, that would mean basically you can't nationalize them because they're international corporations. But if you were to like break that off and say like the government has control over this, could these laws, is that a good political strategy, I guess? is what I'm asking to approach of trying to get these laws to apply to these companies. If we nationalize, soren, does that make it okay? Yeah, no, I guess I wouldn't go as far as saying we should nationalize them because I have other feelings about what we should do, which is not to enshrine them in our permanent system. But I think that if we consider them state actors, so sometimes in some situations,

when a company or when an entity becomes so entwined with the government or works so closely with the government. There are a few theories that this works that that are applied to these situations. But there are certain situations where the government consider or a non-government entity is considered a state actor because it works so closely with the government or it does such a large part of a government's work that it is then required to follow, you know, the same constitutional obligations that an agency would have to follow. Like maybe LexisNexis is so intertwined, with the work of ICE surveillance, that it should be treated like a part of ice for the purpose of, you know, warrant requirements, et cetera, et cetera.

Yeah, and that brings me to the next question, which is, is Batman a state actor? It depends. Yeah. So this is from actually, I think it's a Law Review article. It's a really fun law review article that I talk about in the book that is not advertising the book. Not hyping the book. I'm just saying that if you wanted to know more about that article, it happens to be discussed in the book. But yeah, it's, it's, it's tough, it's a tough call.

It's, it's one of those, like, here's the lawyer's answer is it depends. But could a Superman, who is doing the work of policing oftentimes for the city of Metropolis, which is a government entity, it's a public, it's a public institution is when Superman fights crimes and gets the bad guys, gets the villains, is he a state actor? And the answer. answer is possibly, right, depending on how intertwined his work is with the government and whether he is doing like a significantly large portion of a government duty. Yeah, I think, I think Superman could potentially be a state actor. And I definitely think that LexisNexis, when doing public institutions work for them, could be considered a state actor that needs to follow warrant requirements and other constitutional requirements. Also, if you're just doing any kind of cop shit, I think that counts. Yeah. Yeah. And I also, so it's, it's, there's, there's a fourth amendment that requires you to get a warrant before you search people's stuff, including their data. But there's also

this law. I have this whole soapbox. There's this law called the Privacy Act of 1974. And the reason it was passed in the 70s is because that's when kind of technologists realized that soon the government would be able to collect massive amounts of data about people and store it and potentially use it for things just like what they're doing today. And the government, you know, Congress was like, whoa, that's, we, we do not want that kind of stuff to happen without the proper safeguards in place. So there's this whole law called the Privacy Act that requires transparency about data programs that the government is doing and limits on those programs. They have to expunge data after a certain amount of time. They have to have like a clear mission and you can't use data

that's for one purpose, for another purpose. Like you can't use somebody's social security number for policing purposes or for other purposes that are beyond that mission. You can only keep them for a certain amount of time. Then you have to get rid of them, et cetera, et cetera. And we don't apply that to today's data programs that we do with data brokers. So I also have trying to be trying to encourage policymakers and lawmakers to think about ways to close. the loopholes in that law so that companies like Lexus have to be more transparent about what they do and put clear limits on what they do. And I don't think that's the ideal fix. Like in my ideal world, these kind of systems wouldn't exist. But if they are going to exist, they should not be as

opaque and creepy as they are right now. Yeah. And so we're getting near time. So I do want to ask, what is your idea of what in a perfect world this would look like in which we have the advantages of digital information, but without the downsides. This is a question I haven't asked in a while, but it's just sort of an experiment and utopian thinking so that we have something to aim towards. And I think that helps make our actual goals concrete because we do complain about a lot of things, but I think this is a fun mental exercise to do once in a while to just be like, what do we want this to look like? That is a really good question. So in my book, which once again, not chilling. I say the way I feel about it is that like there's no one answer. This is such a big

kind of multi-market gigantic problem. Like there are open access issues. There are surveillance issues and they're all being caused by just like the same few companies that have become these massive multi-market information monopolies. So there's not one like you can't there's not one magic law that's going to fix all of the informational problem. But I have like a set. of ideas. And I didn't invent most of these. Most of these are, I just brought them together. But the first is to break up these companies and like not let the same companies that own all of our academic journals also be the nation's biggest data brokers, like maybe make split that up, because that seems messy. And so one is antitrust, right? That that's what that falls under.

It's just breaking up monopolies, making sure these companies stop consolidating and, you know, kind of prevent some of the massive gigantic information barons that we see today. Another is to close loopholes in these kind of state actor rules and make sure that these companies have to follow warrant requirements and have to comply with the Privacy Act of 1974. And that's really focused towards like the creepy policing surveillance stuff. And then as far as public infrastructure goes, I think back in the back in the day, we set up special systems for independent media. So I wouldn't want to like nationalize.

I don't think having national information systems is great when you're like in a really good place with your democracy and with your, you know, neutrality in world. I don't, neutrality is not the word I want to use. I'm not a person who is big on believing that there's neutrality. But just kind of the idea that the government could be independent at all and provide independent news or independent science. But I think what we've seen, you know, in the last while especially is that the government shouldn't be trusted to be our source of information. But what the government can do is the government

can fund public radio, public news, right? Public television. It's done so in the past. It's past laws that have guaranteed and set up and then supported infrastructure that is public, funded by the public, not behold, you know, not under any one gigantic corporate umbrella. but it's also independently run. I would envision systems run by librarians and libraries, people who understand how to organize information, how to ensure intellectual freedom, and they would be systems that could either run parallel to these companies

or be replacements for these companies. But just really what that comes down to is more funding and more support. Because I think one thing that happens a lot is people just say like, I was in a talk a few last month with somebody who basically just said, librarians need to do their job. Like, librarians are just, I can't believe they've abdicated all this power. But, like, we all work in libraries, and we know that libraries are underfunded and understaffed and under, you know, just under cared for.

So I think, really, we have to foster a library environment and library concepts by making sure that librarians are paid. librarians can have the resources they need. There are plenty of us. And I think right now that could use a lot of improvement. Yeah, you mentioned the concept of like a national digital library system. And I was wondering, which was like well staffed with information specialists, similar to the way that like LexisNexis is staffed with lawyers in order to provide like

support and resources immediately. So, but I was kind of confused when you were talking about this digital library. because you put it in the context of open access and open infrastructure. Are you imagining that this library would be hosting and providing the platforms for publication? Or would it be aggregating services in the sort of like national sense? Like we would have a national L-Severe contract. We'd have a national Thompson Reuters contract like India was trying to do. Yeah.

That's so that's funny because the model, yeah, I was actually kind of in my mind thinking about India's model. And also, I think I want to say, like, Mexico and some parts of South America, I don't, I'm saying I'm not sure because I don't know if it's a, like, if it's a multinational consortium or if it's specifically in Mexico, but there are these systems where governments are creating platforms where else of your journal, Springer journals are provided. And that's what I was envisioning, not because I think that right now our peer review editing and journal processes generally are fair or okay. It's more because I wanted to stick on kind of the work that librarians do and really providing that access because I agree. I think that the entire, I don't want to say

obviously, because maybe some people love it. But I think that the whole journal system could use more support, more. There could be changes that would benefit it if it was, or that would benefit kind of the journal system if it was taken out of the oligopoly. But yeah, I was really focusing on just kind of the end user and that access point. Not because I don't think the other points are important, just because I didn't want to open the problem up too broadly, because I think you could write a whole book about how to reform the journal systems and how to make a better journal system that works better for scholars and for end users.

But that's, it's just such a big topic because there are so many problems with the current system. Yeah, I mean, that's not too different from the kind of model that AJ Boston has been advocating, which is the read and let read, which is sort of, a token-based access, like, every public, because, like, the idea of a national contract is pretty far-fetched, so his idea is more like a consortia, like the University of California system would buy access for all of its users, plus two million more, and those tokens kind of get used up by the public as necessary. And I'm sure there would be all kinds of problems in terms of authentication, but it is, like, a way of getting basically, like, these reciprocal borrowing

with the public ideas for academic journals. So, yeah, I can see that working. I am actually kind of a fan of, at least in the short term, meaning like next 10, 20 years, the decentralization of publishing because it breaks up the pieces of the oligopolis, if you can flip journals, even if we're having to run them through the libraries. You mentioned the book that it's not fair to ask librarians to take up all the slack of this. It should be a national program.

and I think maybe something might be happening right now in terms of building national repositories for federal publications. That might be something the Biden administration gets going. But I am also kind of in a break off pieces of it because that's that prestige game. And if you can break them off more and more, then it's kind of worth it in a short term to do that because that's also money that you mentioned gets reinvested in building these data programs, like the money that they make from their. their publishing arms is getting reinvested in these data programs. So the more we can break away from it, the more funding, possibly, we're pulling away from them. I mean, they're probably going to find funding in a million other ways. But it is kind of, I think, a good short-term goal for libraries to get invested in the publishing game in a small capacity,

especially if you can flip an already existing journal. I mean, I work with Texas Digital Library. We have tons of independently hosted journals, and most of us are like, stop making journals. just stop. But if you can flip, because a journal is basically a community. It's basically a podcast. It has fans. It has people who, like, read it.

They are. They are exactly the same thing. Totally, yeah. It has a community of people who contribute to it and people who get our podcast peer review. And people who consume it and people who support it. And that's all community. A journal is a conglomeration of people. It's not a publication, really.

Because if you want to just publish something, you could publish an edited volume. as a book. But a journal is an ongoing project, and that requires lots and lots of ongoing labor and care and participation. So really, you're talking about communities, and if you flip those to nonprofit models, then you're actually moving the whole community away from a for-profit model to a nonprofit model. So that's the way I like to think about it. And I hope I can, I hope that way is more appealing of talking to people than saying, like, these vague terms like open access, taxpayer arguments, those things don't really work. because that's how we got APCs.

Yeah. Yeah. So I came into this. It's interesting because law libraries do things. We have our own system and we actually have a pretty good open system that we make for ourselves just because lawyers don't. Like everybody who publishes in law reviews are lawyers and they don't want to see copyright. So we have our own system.

And I didn't know about all these different types of open access scheme. and I came in like all big and like dove eye like I was like oh this is great and then I realized that like every scheme is just part of the same racket right like there's no scenario where the publishers aren't going to try to profit as much as possible from the system so I'm like okay diamond green gold black I'm like these are all it's because I had I had like a whole chapter in my book where I like took them apart and talked about them and then I just like threw the chapter away and was like you know what these are all ridiculous yeah and you're right they all lead at the end like to this APC thing, like none of it. None of it is a system where a publisher is going to say, you know what, I'm going to take a hit on that. I'm going to make a little less money. There's none of them do that. Yeah.

And I think, yeah. And even the nonprofit publishers, I use it something that made me raise an eyebrow, which was like this data isn't valuable to nonprofits. If you remove the profit motive, it removes the data collection motive. And my eyebrow went up because, like, I think we're still stuck in those logics of everyone's like,

Well, I'm grant funded, so I need to prove engagement. So I still need to track how many people are reading the journals. Yeah. Engagement. Prove engagement assessment is like or like Manchurian candidate. Like sleeper agent. If you reach a tipping point in the in the market, then yeah, people would lose the incentive to do that sort of thing. Yeah.

But I think you all like, I think we all as people who kind of work in this world see that that is a, like you said, it's a really hard ask. You're right. We need kind of interim moves because when is the world that we're going to, where is the world where we're going to just drop prestige where we all agree to like let go at the same time. Okay, if you let go and all millions of us are just like, okay, we're going to drop this antiquated, horrible, racist prestige system. Yeah. That's a tall order. So yeah, I think you're right. Yeah, I see. I agree with you. Yeah, chipping away, I think really is the only thing that makes sense. I mean, again, bar. Varring revolution, you know, I mean, there's really nothing else to say except like, okay, we're just going to like make a long-term goal and chip at it. And it'll reach a tipping point. And then everyone's going to go, oh, okay, we can let go now. And everyone else just has a point where they just, they do the, do the jack thing in Titanic and they just let go and drift off.

I don't know why. That was very vivid. It just came to me. I was like, oh, I haven't seen that movie in like 20 years. It was beautiful. That's beautiful. as like poetry. Yeah, of course.

You can't forget. It's an unforgettable thing. Yeah, gorgeous. Yeah. Yeah. No, I think you're right. And I mean, honestly, it's kind of the same thing with this whole, like, it's taken this long since 2017 for there to be a critical mass of people who are like, wait

a second, maybe read Altheville-Elexus Nexus. This is a data, like, people weren't even willing to kind of make that connection. So it does. It takes a really, it takes a long time. But, yeah, I do think that people are becoming, it's kind of part of critical information literacy as a whole, people are becoming a bit more critical of where our information comes from and kind of what the power structures behind that are. Yeah.

I think there's always going to be like a mainstream sort of liberal approach to information literacy and there's always going to have to be a more radical approach. It's like, okay, but who authenticated information from us and what are the power differentials in there? And that's kind of always the weakness of a basic pluralistic liberal approach to it instead of saying like, no, maybe we should throw a little anarchist theory. in there or something like that just to shake it up and actually challenge our

assumptions. I did want to talk about the, we don't have time, but mentioned just so people might be interested to read a book that we're definitely not promoting about like Lexus terminals and detention centers and the issues there and the thing that really, I tweeted

earlier like I didn't realize Lexus was this evil. So there was a feature in Orange is the New Black where they show a Lexus Terminal in an ICE detention center and how like people can't access it. And then Lexus thought, this is great product placement. Let's talk about how we were featured in Orange is New Black. Aren't we so cool? Isn't that exciting? I'm like, you are a soulless person if you want it. If that was your reaction. Even Arthur looked at that

one and he can't hear you. I've got headphones on. Arthur just sensed the evil in the red How muskified is your brain? That was the most sweet. I got mentioned, dude. Yeah. I should send you the whole because they took down. I think they realized because at some point I pointed out. I was like, well, that is not.

That is in poor taste. And I look, I have it permalink. So I actually still have the original blog post. It is hideous. It is bad. Yeah. They're like, it's so funny that they use their Lexus tournamentals. Meanwhile, there's an actual article.

The reason the producers used it is they went to, on a tour of an ICE detention center in order to prepare. Because the last season of the show had scenes from an ICE detention center because it was around the same. It was around 2017. And they decided to end the show that way. And the reason they included the Lexus Terminal is when they visited that ice facility, they saw all of these, you know, I think, I don't know. I think they went to a women's center. So all of these women, like in, you know, in kind of an enclosure. And then they saw this room off to the side with some computer terminals in it.

That was the quote unquote law library. And it was just some Lexus terminals. Nobody was using them. Nobody, I don't even know if they were allowed to use them. But they, one of the producers had been a lawyer. So they saw the terminals and they were like, wait, that's the only information source these people have. That is, that is just like it seemed, I can't think of the word, but it seems like, a joke almost that that's what they give people who English probably is in their first language.

They know, you know, using a Lexus terminal and reading case law is really complicated, even if you are, you know, even if you, even if you are me or somebody who studies a lot. Like it's complicated stuff, even for experts. So just leaving, leaving these people who are being faced with deportation and permanent separation from their family with a Lexus terminal as their only kind of linked to information that could help them, seems so cruel to the producer that they made a part of this show. And that's what Lexus is dragging about. You're right, Sadie? Moment of silence for my poor enraged heart today. I'm still taking psychic damage from the journal podcast. I'm right.

You are and I hate it. I kind of like it. I think it's kind of. dear. Like, you're right. A journal is like a podcast. It's just, yeah, it's communities. It's ongoing labor and a lot of them die after like one or two issues. But yeah, that also, you also mentioned, because we did a couple issues on, on prison libraries issues. We did a couple episodes on prison libraries. She just flick me off, ban him mods. You mentioned how there's no like secondary law literature that in the, in a print library, there would be secondary literature

that explains how to use the case law. And with a terminal, it's just advanced Boolean searching and a lot of reading on like a janky prison pro tablet, whatever, Securist tablet or something that you can't, maybe in a prison you could take it with you if you had a securest tablet. But like in these detention centers, they were separate glass rooms that you had to go into. Yeah. And one of the, so it's interesting if you look at the way they, they, the legal information companies advertise their quote unquote correctional line of products.

It's not about helping provide access to justice for people who are incarcerated. It's about being able to monitor their computer use. And I actually saw records, I forget what stated from, but somebody was able to FOIA records of people's, how the prison tracks how people use the Lexus kiosks or the Lexus iPads or whatever systems they have. and they could tell whenever somebody tried to go on an unauthorized web page, basically leave. And you can get in trouble for when you're at the kiosk, you know, trying to like leave that web page or that whatever or that URL. And like it's just another way to be punitive, basically, these services. And they're replacing law libraries, which used to be staffed by a human librarian and used to have shelves of books that were, you know, more user friendly and more more kind of diverse and range.

in scope, like instead of just buying like the Florida state case law database, there might be, you know, books from other states or books from, you know, federal books about federal law, books that explain the law that just don't exist as choices for people anymore in prison in jail. Yeah, I'm going to promote the Twitter account. Jail House lawyers speak for people who are interested in, I mean, you can definitely follow people who are tweeting from like contraband phones and stuff like that. But there are a lot of more, there are a lot more resources from the inside than there used to be. And so you should definitely directly hear from people who are affected by incarceration because there's a lot to learn if you haven't been. It's a very different

world. So I am going to wrap up. Sarah, is there anything you don't want to promote? I definitely don't want to promote the book Data Cartel. There definitely will be a link in the notes. By Stanford University Press. No. That would be uncool. Not at all for sure. intense. It would be the most done cool. We have no chill here. Wait. I thought there was also a library juice class that we definitely aren't promoting at all. I should have done

that first and then you end it. That would have been much better. I'm not the fucking guest. But you just devised the whole course. Now the cyber is so big. That's true. I'm been meaning to use that and I can't forget. Yeah. So in March, I will be

teaching Zotero for librarians through Libraries'is Academy. They thought it wouldn't be very popular and almost didn't accept it. So prove them fucking wrong. There's 30 seats. You can register through the end of the first week. It's $200. Get your institution to pay for it. And then I get money and it's great.

It won't, it's not just like, oh, hey, learn Zotero. It's more focused on like, okay, learn Zotero so that you can like make resource guides and teach patrons and customize it to fit. your needs if you're like writing a grant proposal and stuff like that. So it's more than just here's how you put things in there and then write a paper with it. Yay. I mean, not yay. That sounds, no. Yeah, boo. You suck. Because definitely not promoting it, boo. No. I hope, I hope that there's overflow. I hope, yeah. Don't you know, Jay's on library punk? Yeah. I'm fucking cool. Do you fucking know who I am?

She's fucking know I am, bro. I hate that this happened right as Twitter's exploding because I got like reach on that platform among the librarian crowd. And now everyone's like, oh, I'm going to Mastodon. I'm like fucking fine fucking nerds, okay. Yeah, well. I'll still be there. I had a Mastodon for like a year. I'm watching people deal with their Twitter addiction in real time.

It's like, these isn't a real community. You don't need 2,000 people to read, taking a shit and farting, LOL. That's not a community. You can join like a forum with 100 people, and those can be your friends. Like, this is not community. It's fine.

You're just addicted to Twitter. It's fine. It'll be fine. We've replaced all of our real friends with Twitter. I'm being very Buddhist about it, like all these people are deleting their accounts and I might never talk to them again and all life is impermanent. And we're all on Tumblr.

You can find us. Yeah. At least never left. That's why we're like this. All right. Thanks, Sarah, for coming on. This is awesome. I enjoyed the book.

It was good to meet all of you and talk to you. Yay. Yeah. Yeah. Yeah. Whatever you want. Good night.