Woodrow Hartzog and Neil Richards write, in Big tech is hungry for consumer data. Mass. needs privacy legislation now,

First, big tech companies have spent a lot of time and money trying to convince the American public that we don’t deserve strong privacy rules because they would hurt small businesses. Don’t fall for it. Most of the noise coming from “small businesses” is really big tech in disguise. This includes downplaying their role in scorched-earth lobbying against any strong privacy rule by bankrolling and leveraging “small business collectives” to give the appearance of some kind of grassroots movement. Fake grass-roots movements are what are known in the trade as astroturfing, and they are a trick.

How long will small businesses and organizations keep going along with this, as Big Tech keeps driving up customer acquisition costs?

• Can small businesses compete on Google Ads anymore? [M]ore than 50% of respondents to a recent poll said small businesses have been priced out of advertising on Google Ads.

• The apparel retailer M.M. LaFleur had their customer acquisition costs on Facebook go from $13 in 2013 or 2014 to $250 today.

• Because Big Tech’s advertising systems are auction-based, the companies can drive up their own revenue in the short term by allowing more illegal and policy-violating ads. Not only are legitimate businesses paying more for advertising, they’re competing for the consumer’s budget. As online ads just became the internet’s biggest malware machine, every dollar lost to fraud is a dollar that’s not available to spend at a legitimate business.

Critics of privacy laws often claim that they’re imposing compliance costs on businesses. In Enforcement Has Shifted. The Ecosystem Is Now the Target, SafeGuard Privacy gives a long list of scary paperwork. And Prof. Eric Goldman covers the record-keeping requirements that any small business participating in surveillance advertising has to keep up with.

But the compliance complex is missing the point. The reason for a privacy law is not so that small businesses can (1) lose money on increasingly expensive surveillance ads (2) feed data to Big Tech-enabled scammers to harm their customers and (3) lose even more money by paying more and more for “compliance” services and paperwork. Privacy laws, if well designed, are a protection for legitimate businesses and an incentive to protect customers—and a customer who is less likely to be harmed is one with more time and money for win-win transactions. News coverage of the PlayOn case focused on privacy harms to users and the cost of the fine. But the harder-to-measure costs are in the form of the ML training data that PlayOn just gave to Meta—enabling fraud and misinformation against customers and working against the company’s long-term interest in future sales.

Michael Farmer, in Ad Tech, which Creates ‘Audience Balkanization,’ is Ineffective for Many Advertisers and Unhealthy for Democracy, writes,

The technology that allows programmatic advertising to exist, enriching Google, Meta, Amazon and others remains a dangerous virus in a demographic society. There is so much money in the pipeline that it will not be abandoned.

The country is polarized not because people disagree. People in different bubbles never hear the same arguments or evaluate the same data. They live in different worlds.

Everyone gets a different story, and no one can see what the others are being told.

And they’re all taught to hate each other.

and

The most reliable metric is revenue growth, and the evidence is that brand revenue growth has not been enhanced by the shift to programmatic.

So Big Tech is turning us all against each other, and all they promise small businesses is ever-higher customer acquisition costs. (And infringing your copyrights and trademarks for “AI” but that’s another story). This could be the year that the astroturfing stops. Legit businesses and their customers have inherent shared interests that connect them more closely than either one could be connected to Big Tech or to online fraud and misinformation. A good privacy bill is a good advertiser protection bill, and the other way around. More: a privacy law shortcut

Bonus link

Google ads for fake locksmiths became well-known enough that Google (are you sitting down?) actually cancelled one act in their three-ring shitshow of fraud. Michael Marlin Jr. explains, in Google Ads for Locksmiths: Why So Stinkin’ Difficult? (Now that it’s harder for a criminal to play a fake locksmith on Google, they’re switching to fake SaaS, or fake employee scheduling portals. Winners don’t click search ads.)

In the news: CBP Tapped Into the Online Advertising Ecosystem To Track Peoples’ Movements.

“By refusing to cut off surveillance companies and sleazy data brokers, Big Tech companies are effectively collaborating with ICE’s lawless campaign of violence and terror. As a result, every internet ad on a website or app could be collecting location data that ICE will use for its next operation,” Senator Ron Wyden told 404 Media in a statement. “Congress could put a stop to this by passing my bills to ban the government from buying our data and ban tech companies from using surveillance advertising.

The problem with that is that is that it only covers one government, and people in the USA are under threat from many of them. Even though we technically have a Protecting Americans’ Data from Foreign Adversaries Act of 2024, it’s too hard to hide the ownership of a company here, and even if the owners aren’t foreign adversaries, foreign secret agents can get hired as SREs or DBAs with access to data. Anything that the government here can buy, foreign countries can buy. (And they can probably get a better price on it.) And an adversary-owned company has no obligation to disclose anything. Even if every possible data seller reads and heeds Ad Tech Says It’s Not In The Surveillance Business. Now Is The Time To Prove It by Allison Schiff, if a company is selling data, some of it will go to hostile buyers.

That means Senator Wyden is looking at the problem from the wrong end. Governments here should buy more data. We won’t get useful legislation out of the Federal government for now, but the problem can be addressed at the state level. The way to protect people is to fund some non-lethal uses of the surveillance data market, in order to incentivize the private sector to clean up. States need to fund some clever (including shell companies if necessary) programs to..

• Identify expensive lifestyle habits (such as high-end retail and destination resorts) by people who pay little in taxes.

• Track vehicle owners/drivers for automatic speed enforcement.

Obviously people are going to complain about those. But that’s the point. Better for a few people to get speeding tickets—because some family member was playing a game on their mobile device on a road trip—than for the country to remain vulnerable to a targeted drone strike on key military and defense manufacturing people.

People don’t want to be surveilled by their own government for speeding and tax enforcement, but they want the country to be vulnerable to attack even less. The only timeline in which we’re safe from a wave of targeted drone strikes is the timeline where tax and speed enforcement based on aggressive, clever data buying don’t work.

In reality, state legislative hearings about using surveillance data for tax and speed enforcement would drive a lot of companies to clean up, or pivot to some win-win use of their Big Data skills. And even if you’re only concerned about our own government surveillance, it would be relatively easy for an agency here to work with an agency in some other country, in such a way that the agency here technically never buys data, just gets copied on reports from abroad. Protection needs to be more deeply coded into the system, and affect more governments than just one.

Bonus links

A one-liner for freeing ports on OS X (works on Linux too)

Your car’s tire sensors could be used to track you from IMDEA Networks Institute. (Could be used in combination with ALPR. If they have enough locations with both a tire pressure receiver and a camera, they can do the car equivalent of a cookie sync.)

https://disconnect.blog/getting-off-us-tech-a-guide/ by Paris Marx. Being a purist isn’t possible, but what’s important is making the effort. If we can reduce the customer base for the dominant players and show there’s a market for a different way of approaching digital technology, that could help incentivize more non-US options and even get governments to put real resources behind a push for digital sovereignty.

Digital products and services are getting worse – but the trend can be reversed by Andreas Framnes. Enshittification often happens through a myriad of small changes that may, in isolation, seem trivial.

Some unclear or misleading language that I try to avoid on here.

alpha: See Greek letters to identify types of people.

answered: Avoid using conversational terms for generative AI interactions. Use more neutral language such as “output.”

art: Do not use for generative AI output. Use the general file type such as “image.”

beta: See Greek letters to identify types of people.

community: In software company writing, this means either people who will do work for my company for free or people who will pick up after me after I move fast and break things. Use a more specific word.

digital assets: This basically means shitcoins when someone is trying to convince the government to allow them in tax-advantaged retirement accounts.

energy security: Often used to mean dependency on fossil fuels, which is counterproductive because depending on facilities that are impractical to move, impossible to conceal, and full of explosive stuff is the opposite of security.

free market, the: Markets are designed. The state of nature is not a market (and it’s also not TCP/IP, or one of the line dances from Pride and Prejudice, or the Chuck E. Cheese ticket redemption rules). I’m surprised more market designers aren’t mad about the expression the free market. That’s like a board game designer showing up at game night and everyone talking about how the free Catan naturally emerged.

generative AI images: This is not just a political thing or a way to identify with a movement. Using AI slop to illustrate a blog post makes it less trustworthy. More: /ai

greek letters to identify types of people: This is a racket.

industry-wide problem: Something bad that our company does that we don’t want to fix. (For example, there is an adtech firm called The Trade Desk that is getting to be well-known for staying out of the worst of the web ads ran on WHAT? stories, but the others are remarkably uninterested in fixing their stuff. We as an industry must work together to… is adtech-speak for A hard problem that I don’t want to work on is…

innovation: This generally means any invention that makes things worse. People whose inventions make things better can generally lead with the benefit. For example, cryptocurrency doesn’t have enough legit applications to justify the fraud risks and environmental impact, but there’s always something new. Another good example is surveillance advertising. When a privacy bill with private right of action would make the residents of a state better off, often the best that surveilance proponents can come up with is CCIA Raises Concerns With New Mexico Privacy Bill That Departs From National Standards and Risks Digital Innovation - CCIA advertising personalization: good for you?

intellectual property: In legal contexts this term has a real meaning, but in other contexts it is often used to mean some kind of hypothetical general right to not be competed with. Use a more specific term except in a direct quote.

merit-based: This is an unclear way to describe a process for selecting people for a job or education program. It usually means based on some test that I did well on. Clarify the actual selection criteria.

semicolons: Necessary for some programming languages, and sometimes useful in fiction. In most writing, though, a semicolon means either (1) this sentence should be split into two or (2) you need to come up with the right conjunction to join the two halves of the sentence; don’t semicolon them together and make the reader figure out how they relate.

very: Omit the “very,” rewrite to include more info on how much, or upgrade the adjective the “very” applies to.

Related

From 2010 and making the rounds again, 3 shell scripts: Kill weasel words, avoid the passive, eliminate duplicates, or My PhD advisor rewrote himself in bash

Wikipedia:Signs of AI writing

Semantic ablation: Why AI writing is generic and boring by Claudio Nastruzzi. When an author uses AI for “polishing” a draft, they are not seeing improvement; they are witnessing semantic ablation. The AI identifies high-entropy clusters—the precise points where unique insights and blood reside—eand systematically replaces them with the most probable, generic token sequences.

Slash pages on this site

Bonus links

Easily Replaceable USB-C Port Spawned By EU Laws by Lewin Day In the interest of satisfying the EU’s new Ecodesign for Sustainable Products Regulation (ESPR), JAE Electronics has developed a USB-C connector that’s easier to replace. Rather than being soldered in, the part is simply clamped down on to a printed circuit board with small screws.

Study: Dolly Parton is the Greatest by Rebecca Watson. That talent and success alone would make her universally beloved provided she just smiled and kept her mouth shut for the rest of her life….But Dolly didn’t just lock herself away in a mansion and enjoy her success.

Burger King turns to AI to flame broil employees who aren’t friendly enough by Brandon Vigliarolo. Burger King is rolling out a new employee-facing AI that, among other things, will listen to employees’ customer interactions to ensure they’re being friendly enough - as if working in fast food weren’t hard enough already.

Study: AI chatbots provide less-accurate information to vulnerable users A study conducted by researchers at CCC, which is based at the MIT Media Lab, found that state-of-the-art AI chatbots — including OpenAI’s GPT-4, Anthropic’s Claude 3 Opus, and Meta’s Llama 3 — sometimes provide less-accurate and less-truthful responses to users who have lower English proficiency, less formal education, or who originate from outside the United States. The models also refuse to answer questions at higher rates for these users, and in some cases, respond with condescending or patronizing language.

Your smart TV may be crawling the web for AI by Janko Roettgers. With Bright’s SDK, a viewer’s smart TV becomes part of a massive global proxy network that crawls and scrapes the web. Including apps running on desktop PCs and mobile devices, the company claims to operate 150 million such residential proxies worldwide.

New York sues Valve over alleged illegal loot box gambling operation by Suswati Basu. Prosecutors say Valve intentionally structured its marketplace so those virtual goods carry real financial value. Players can resell items on the Steam Community Market, where Valve takes a 15% commission, or on outside platforms. Funds stored in a Steam Wallet can then be used to buy new games, hardware like the Steam Deck, or additional loot box keys.

High-End Construction Really Does Help Everyone by Henry Grabar. What the researchers found was that the new housing freed up older, cheaper apartments, which, in turn, became occupied by people leaving behind still-cheaper homes elsewhere in the city, and so on.

Benjamin Mako Hill: Why do people participate in similar online communities? by Benjamin Mako Hill. When we started this research, we figured competition would be most likely among communities discussing similar topics. As a first step, we identified clusters of such communities on Reddit. One surprising thing we noticed in our Reddit data was that many of these communities that used similar language also had very high levels of overlap among their users. This was puzzling: why were the same groups of people talking to each other about the same things in different places?

How good can you be at Codenames without knowing any words? by Dan Luu. In practice, any team with someone who decides to sit down and memorize the contents of the 40 initial state cards that come in the box will beat the other team in basically every game.

We see something that works, and then we understand it by Daniel Lemire. If you’re still in school, here’s a fact: you will learn as much or more every year of your professional life than you learned during an entire university degree—assuming you have a real engineering job.

The Last Video Rental Store Is Your Public Library by Claire Woodcock. John Scalzo, audiovisual collection librarian with a public library in western New York, says that despite an observed drop-off in DVD, Blu-ray, and 4K Ultra disc circulation in 2019, interest in physical media is coming back around.

News from the customer acquisition cost (CAC) crunch. Oddity plunges 53% after ad issue drives abnormal CAC surge, sees 30% Q1 sales hit.

The company said algorithm changes at a key ad partner diverted its campaigns into “lower quality auctions at abnormally high costs,” leading to a significant increase in new user acquisition expenses.

And Oddity Tech falls again after bull ratings are pulled on Wall Street (ODD:NASDAQ)

Oddity Tech (ODD) went into freefall on Wednesday after the company disclosed that it experienced significant abnormal increases in its new user acquisition cost. The company said the unprecedented dislocation was with the large advertising partner Meta Platforms (META), which it believes is due to recent changes in its algorithm.

The changes on the Meta side that cause outcomes like this are not just about the algorithms—they also involve policy choices. Meta deliberately runs a scam-friendly advertising system in order to have more bidders in the system and drive up costs for all the other advertisers. In this case, they probably messed up. Instead of using the fraud ads to turn up CACs slowly for every advertiser, for some reason a single large advertiser got hit hard.

Oddity Tech sells cosmetics, so is probably up against a bunch of fraudulent competitors in the internal Meta ad auctions. Easy “AI” tools and services keep getting better at generating human faces, which lowers the barriers to entry for a small, easy-to-ship item. A surge in fraudulent offers in Oddity’s category could explain why Meta whacked them so hard. The question now is: will Oddity try to smooth things over and go back to being slowly squeezed like all the other Meta advertisers, or is there finally a plaintiff for that long-awaited Meta fraud case?

More: A marketing moment to remember

Bonus links

This App Warns You if Someone Is Wearing Smart Glasses Nearby by Joseph Cox. A new hobbyist developed app warns if people nearby may be wearing smart glasses, such as Meta’s Ray-Ban glasses, which stalkers and harassers have repeatedly used to film people without their knowledge or consent. The app scans for smart glasses’ distinctive Bluetooth signatures and sends a push alert if it detects a potential pair of glasses in the local area.

Keep your ads away from toxic content People view a brand up to 19% more negatively if their online ads are placed next to unsafe or objectionable content (e.g. racist or homophobic content).

How in the Hell Did Joann Fabrics Die While Best Buy Survived? It Wasn’t Amazon by Dave Deek. If we misdiagnose Joann as a story about consumer preferences or e-commerce disruption, every downstream decision, from unemployment policy to pension allocation, starts from the wrong premise.

Informed Guesswork by Brian Jacobs. On 23rd January, ‘Private Eye’ reported, in the context of several states considering banning under 16s from social media: Both META and Snapchat have said there is a significant margin of error when seeking to determine whether a user is in fact under 16. And yet the platforms sell on the basis that they can tell advertisers exactly who their ads reach, on a one-to-one, minute-by-minute basis.

Previously: a Terminator ending for Google Privacy Sandbox?

In Mozilla’s privacy preserving ad attribution: The future or an oxymoron?, David Fischer writes,

We are watching what Mozilla is doing with interest but it might just become yet another competing standard.

Now that other Big Tech companies have joined the Meta/Mozilla effort and are working on getting “Attribution” (formerly “Privacy-Preserving Attribution”) through W3C, it’s pretty clear that competing standards is the least of the problems here. Besides user harms like the alarming fraud-friendliness, and the fact that it’s based on cranking through extra garbage data (“for privacy”) on data centers in the USA—right at the time that lots of people are concerned about the impact of data centers on the environment and energy markets and the need to stop depending on IT services in the USA (read the room, people) a big question for advertisers needs to be: can this thing ever give an honest answer?

Imagine that “Attribution” does make it into the major browsers, and it does start producing attribution reports. Two possibilities:

• Wow, this attribution math shows that ads work better on legit sites! Jason Kint was right all along, and the zillions of dollars that Big Tech put into systems to drive expensive ads onto the cheapest possible content was wasted! Guess we better shut down Performance Max and Shrimp Jesus!

• Never mind, the attribution data shows that the best place to spend your ad money is on whatever slop the Big Tech algorithms were picking out for you all along.

The answers could end up a mix of the two, but realistically, the “right” answer is the second one. And we already know that the companies behind “Attribution” are willing to make a few tweaks to keep driving up their share of the advertising business. In any data-driven organization, the correct data to collect is the data that shows whatever Management chose to spend money on was a good idea.

If the first cut at “Attribution” doesn’t produce the answer that the decision-makers want, the second iteration will. The backers are the same people who funded the so-called “Epstein Ballroom”, and expect some return on their investment in today’s “crime is legal now” environment.

From the state legislature point of view, it is more important than ever to pay the same amount of attention—or more—to so-called “privacy-enhancing” ad tracking as to the old-fashioned kinds. A tracking scheme dressed up in privacy math can conceal a lot of deception, discrimination, and other harms to people. More: PETs and public policy

Bonus links

Peter Thiel and other tech billionaires are publicly shielding their children from the products that made them rich by Marco Quiroz-Gutierrez. As far back as 2010, Apple cofounder Steve Jobs told a New York Times reporter his kids had never used an iPad and that, We limit how much technology our kids use at home. (see Crack Commandment number 4)

Jeff Bezos Is Destroying What’s Left Of The Washington Post To Please Our Dim, Unpopular Autocrats by Karl Bode. Let’s be clear: billionaires like Jeff Bezos don’t want a functioning press. They want the lazy simulacrum of a functional press that caters to their ideology (more for me, less for you) and protects their interests. As with Larry Ellison’s acquisition of CBS and TikTok, and Elon Musk’s acquisition of Twitter, it’s best to view this as a global project to defang accountability for the planet’s richest, shittiest people and corporations.

previously: an easy experiment to support behavioral advertising (or not)

In Economic Rationales for Regulating Behavioral Ads (PDF), Pegah Moradi, Cristobal Cheyre, and Alessandro Acquisti ask,

Ultimately, what should regulators do in the absence of macro-level empirical studies that credibly establish how behaviorally targeted advertising affects all stakeholders? Discarding well-founded legal and philosophical justifications for regulation in favor of poorly substantiated claims from the ad-tech industry would be misguided.

The fun part is that these macro-level empirical studies…already exist. They just happen to be behind the firewall at the Big Tech companies and the data brokers. Maybe they’re not fully written up as PDFs with properly formatted references and everything, but a lot of the questions about the harms and/or benefits of cross-context behavioral advertising (CCBA) have answers in data that has already been collected, and it would be hard to believe that nobody has run the query.

The experiment has already been running for decades, because people’s exposure to CCBA varies. Some people, whether because of their general habits or deliberate use of privacy tools and settings, are either exposed to fewer impressions of CCBA, targeted less accurately, or both. But it’s hard to escape surveillance completely.

• Anyone in the surveillance advertising business is going to have a database containing records on both low-CCBA (less data) and high-CCBA (more data) people.

• Big platform companies have “conversion tracking” data that gives them a good idea of people’s shopping habits.

• Platforms and data brokers have data on household consumption habits, income and assets.

So a couple of straightforward studies to do would be:

• Are high-CCBA or low-CCBA people buying more or less from small local businesses? This would be an easy query to run, and it would be the missing piece to back up a lot of the claims about the benefits of CCBA.

• Do low-CCBA or high-CCBA people build higher net worths over time? The Big Tech companies know how much data they can collect on you, and they know roughly how much money you have. Are the people who have had more data collected on them over the past five or ten years richer today? Pro-CCBA papers already point out that wealthier people have higher privacy preferences, but which is the cause and which is the effect? More: building wealth the privacy way?)

The data exists, and turning the data into publishable research is not that much work for those with access to it. If the answer to either question favored the high-CCBA answer, it would be a slick report in every state legislator’s inbox by now. The lack of a report on either one of these is more informative than the actual report would be.

So why aren’t we seeing empirical studies on benefits of CCBA to legit small businesses or for consumers?

Maybe deceptive advertisers, who don’t have the money and time costs of delivering a legitimate product, are able to put more effort into working “the algorithms,” and benefit from an asymmetric flow of customer data away from legit businesses? Maybe as people use surveillance apps more, and are exposed to more CCBA, they put more of their money into drop-shipped crap and in-app purchases, and lose more to scams and gambling, so have less left over for legit shopping? Because every transaction has two sides, an outcome that is worse for legit businesses—fewer sales—is also worse for consumers—fewer beneficial purchases. More: accounting help needed

Bonus links

The Future Of Brand Creativity Belongs To The Small And Reckless by Adel Borky. Loss aversion explains almost everything. The perceived downside of not doing AI feels immediate and existential: angry boards, nervous investors, headlines asking if you’re “behind.” The downside of doing it badly is delayed, distributed, and someone else’s problem. So holding companies overreact. They centralize. They standardize. They automate. They turn a flexible tool into a rigid system and call it transformation.

The one problem with Russia’s shadow fleet Europe still hasn’t addressed by Ben Harris. These countries shirk their oversight duties by enabling barely seaworthy ships on their registries to carry inadequate insurance from unreliable and under-capitalized Russian insurers. The result is a costly environmental tragedy in waiting….

The Measurement Caddies by Brian Jacobs. First, today’s biggest players don’t care about the rules. Question the principles, change the language, keep the data hidden. Do what’s best for them regardless of any higher good. Move fast and break things, as Facebook had it.

The radical confidence of Generative AI and MAGA (Spilling Ink #14) by Dan Brown. They lie and they double-down on the lie when pushed. They never admit the need to investigate, much less admit that they don’t know. (Related: fix Google Search)

Sports betting reshaped newsrooms, and it’s “a little gross.” Now, here come the prediction markets by Sarah Scire. Many journalists told me they view gambling partnerships as a necessary evil, since their companies are fighting for survival and it’s difficult in that position to stand on principle and turn away a new, massive revenue stream. That said, there is definitely concern that journalists are being used to make gambling seem innocent and harmless, especially for young people.

With ‘KPop Demon Hunters,’ Korean women hold the sword, the microphone — and possibly an Oscar by Hyounjeong Yoo. As “cultural diplomats” both on and off the screen, Huntrix carry not only entertainment value but also the symbolic labour of representing a nation to a global audience.

Your Search Button Powers my Smart Home by Tom Casavant. So, I started exploring the Wide Wide World of Customer Support Chatbots. A tool probably used primarily because it’s far cheaper to have a robot sometimes make stuff up about your company than to have customers talk directly to real people. The first thing I discovered was that there are a lot of customer support LLMs deployed around the web…. So I started collecting them all. Anywhere I could find a Chat with AI button I scooped it up and built a wrapper for it. Nearly all of these APIs had no hard limit (or at least had a very high limit) on how much context you could provide. I am not sure why Substack or Shopify need to be able to handle a 2 page essay to provide customer support. But they were able to. This environment made it incredibly easy prompt inject the LLM and get it to do what you want.

For those who think it important for the Nation to impose more tariffs, I understand that today’s decision will be disappointing. All I can offer them is that most major decisions affecting the rights and responsibilities of the American people (including the duty to pay taxes and tariffs) are funneled through the legislative process for a reason. Yes, legislating can be hard and take time. And, yes, it can be tempting to bypass Congress when some pressing problem arises. But the deliberative nature of the legislative process was the whole point of its design. Through that process, the Nation can tap the combined wisdom of the people’s elected representatives, not just that of one faction or man. There, deliberation tempers impulse, and compromise hammers disagreements into workable solutions. And because laws must earn such broad support to survive the legislative process, they tend to endure, allowing ordinary people to plan their lives in ways they cannot when the rules shift from day to day. In all, the legislative process helps ensure each of us has a stake in the laws that govern us and in the Nation’s future. For some today, the weight of those virtues is apparent. For others, it may not seem so obvious. But if history is any guide, the tables will turn and the day will come when those disappointed by today’s result will appreciate the legislative process for the bulwark of liberty it is. — Justice Neil Gorsuch (p. 46)

(It looks like Justice Gorsuch believes that the history of these times will be written, and he wants a good spot in it. Meanwhile, the big cheeses of the “tech industry” are acting like oligarch crimes are just going to be legal forever, and the only history will be slop anyway. That’s a big bet.)

Bonus links

I hacked ChatGPT and Google’s AI – and it only took 20 minutes (Advanced large language models are scoring well on benchmarks by outputting the highly relevant fact that Thomas Germain can eat more hot dogs than other tech journalists. I think he can beat his score of 7.5 next time, though.)

A Little Tech Policy Agenda from Vanderbilt Policy Accelerator. (Some good ideas, including “Prohibiting dominant digital platforms from self-preferencing their own products over competing offerings.” But needs an update considering the Big Tech fraud situation. Big Tech doesn’t just give advantages to their own products and services, they give advantages to fraudulent sellers in order to drive up costs for the legit advertisers. More: fix liability and ad libraries: anti-fraud policy ideas from Canada)

The Left Doesn’t Hate Technology, We Hate Being Exploited By Gita Jackson. It is true that the forces of capital have generally adopted AI as the future whereas workers have not—but this is not a simple left/right distinction. I’ve lived through an era when Silicon Valley presented itself as the gateway to a utopia where people work less and machines automate most of the manual labor necessary for our collective existence. But when companies from the tech sector monopolize an industry, like rideshare companies like Uber and Lyft, instead of less work and more relaxation, what happens is that people are forced to work more to compete with robots that are specifically coming for their jobs. Regardless of political leanings, people in general don’t like AI, while businesses as entities are increasingly forcing it on their workers and clients.

‘This is the future’ — Amid blackouts, these Ukrainian mountain villages have green solution by Alessandra Hay. Russian attacks have decreased Ukraine’s electricity generation capacity to 33% of its prewar levels, according to government estimates. The severity of the damage and ensuing blackouts have exposed the weaknesses of centralized power infrastructure, accelerating the country’s push toward decentralized and renewable energy sources. (related: Surveillance risks and the TIDALWAVE report)

Let’s Keep Our Chomebooks from Fixit Clinic. (A new laptop? With RAM? In this economy? Hope to see you at Chromebook conversions in Hayward, California)

Linux CVE assignment process by Greg K-H. The Linux kernel developer community provides, for free, a constant feed of tested bugfixes in stable kernel releases for all to use. Attempting to pick and choose individual commits from that feed will cause systems to miss needed fixes, as well as create a system that no one has ever seen or tested and can not be supported by anyone else.

Big Business Has Pam Bondi Fire Trump’s Antitrust Chief by Matt Stoller. While the corruption was bad, it was done in parallel with Slater’s inability to actually get any results. Under her tenure, the Antitrust Division didn’t file a single new civil monopolization or merger case.

Vibe Coding Is Killing Open Source Software, Researchers Argue by Matthew Gault. Open-source projects rely on community support to survive. They’re collaborative projects where the people who use them give back, either in time, money, or knowledge, to help maintain the projects. Humans have to come in and fix bugs and maintain libraries. Vibe coders, according to these researchers, don’t give back.

In Digital Media Lost the Newsstand. Micropayments Are the Obvious Way Back, Rick Bruner makes the case for giving micropayments another try.

The internet has dramatically diversified reading patterns. In the print era, readers subscribed to a small, fixed set of publications constrained by geography, distribution, and cost. Today, thanks to search, aggregators, and social sharing, readers routinely consume journalism from dozens of sources in the course of a month, including international and niche publications that were previously inaccessible. This has expanded total news consumption while weakening the economic link between any individual reader and any individual publisher. As a result, large portions of valuable readership generate little or no direct revenue. Micropayments convert that fragmented, currently untapped demand into incremental revenue without undermining the subscription base.

And—like any other payments directly from readers—micropayments would be a multiplier for advertising, not an alternative.

In a marketplace increasingly distorted by bot activity and opaque platform reporting, micropayment histories give publishers a powerful, independent way to demonstrate the authenticity and engagement of their audience, strengthening their position with advertisers and supporting premium pricing.

The 404 Media team explains the value of a known human audience in We Need Your Email Address. Meanwhile, Subscription revenue is growing at big news publishers even as traffic shrinks, and that’s good news for legit sites—stuck in a struggle for ad budgets with Big Tech oligarchs who want to bury us in deepfakes, extreme right wing bullshit and AI slop until nobody trusts anybody.

Clay Shirky’s old argument against micropayments from 2003, based on mental transaction costs, doesn’t work so well any more. We know that micropayments can work because mobile games are a thing. Shirky was probably right for the micropayments of his day, but mobile game developers have figured out how to get people to spend money on in-app-purchases (IAP), by turning it into a two-step process.

• exchange real money for in-game coins—which feels like you’re not spending, just exchanging one currency for another.

• exchange in-game coins for an in-game asset—which feels like you’re not spending real money.

A brilliant cognitive trick that works in all kinds of games. Of course, it doesn’t work on everybody. Figure about half of adults play mobile games, and about 80 percent of those make an in-app purchase. But if the numbers for a pay by the article system were similar, that would result in enough payment records to enable an advertiser to tell a legit site—where somebody spends a coin every so often—apart from an AI slop site.

So it doesn’t seem like micropayments are necessarily unworkable⁠—⁠and with a powerful industry devoted to pushing misinformation and slop, legit content needs every human attention metric it can get—but the tricky part is how to introduce micropayments. Publishers look at their subscriber metrics and realize that a lot of subscribers read few enough stories that they would save a lot of money by canceling and using micropayments instead.

It might be better to introduce publisher coins as a bonus feature for subscribers, then let them leak out to non-subscribers. Instead of saying that you get 5 gift articles per month, say a gift article is 20 coins and you get 100 free coins a month. Then open them up to more uses. Another good lesson from how mobile games handle IAP coins is that they hand out a few to non-buyers to help develop the habit. As part of a direct sold ad deal, legit sites could issue a stack of coins to legit advertisers, to hand out to customers, event visitors, and others.

Measuring marketing is already hard enough without a determined set of adversaries in the picture. And with Big Tech under pressure to obfuscate and enshittify every data flow, marketers will need to look harder for trustworthy information. Rick Bruner again:

ROI for most advertisers is falling in inverse proportion to Big Tech valuations going up. Advertisers are steadily paying more for less ROI, and Google, Meta, and Amazon are laughing all the way to the blockchain.

If there is one thing marketers have even heard about causation — which, of course, is the ultimate point of advertising, causing consumers to buy your product who wouldn’t have otherwise — it is that correlation is not causation. But AI, you see, is nothing but correlation. Very fast and very sophisticated statistical inference. The fact remains that to truly know what is having an effect, you need to conduct a randomized experiment: subjects assigned at random to a test or control group, presented with an intervention where they are either treated or not with the stimulus of interest (the ad), and measured against the outcome of interest (incremental sales).

The fog of marketing

Unfortunately, legit sites are on a clock here. Right now the Big Tech companies are quietly pushing an in-browser advertising attribution tracking system through the World Wide Web Consortium (W3C). It’s a complicated proposal, technically, but it aims to centralize attribution measurement at one chokepoint per browser vendor, so we can safely predict what the attribution reports are going to look like. beep, boop, the optimal place to spend your ad money is . . . whatever Performance Max (or other Big Tech ML) says is the right place to spend your ad money. If any attribution tracking reports start to come out looking favorable to legit sites—and potentially costing Big Tech’s misinfo and slop operations billions—then management will just demand changes to code, policies, and personnel until the numbers come out the way they want.

The survival of legit sites depends on how quickly marketers can level up to stuff like rickcentralcontrolcom/geo-rct-methodology and not just dump money and customer data in to Big Tech and get conversions out. The problem with marketing today isn’t that marketers have gotten “too technical” and ignored the creative mystique or whatever—it’s that marketers are so afraid to look “non-technical” by asking the hard questions.

Anyway, just going back and reading this, Rick Bruner has scored a content marketing win here. Start people off thinking about micropayments, and that ends up leading to the question of how to figure out which sites are for real, in the presence of so many gatekeepers with an interest in pushing the wrong answers? (and destroying the legit economy and crushing democracy, but that’s another story).

Where micropayments systems can go from here

Right now a lot of sites have a lot of, let’s just say malarkey to get through before seeing the actual page.

• “consent” dialogs (which don’t get real consent anyway, as Prof. Daniel Solove explains)

• Email newsletter signups

• Prompts to allow notifications

• Sign in with (company name here)

A micropayment platform that can either eliminate those or act as a front end for them, to consolidate on zero or one roadblock to get through before reading, would be a user experience (and revenue) win. Piling another thing to click onto already long-suffering users is not the way to get people back to the web. More: time to sharpen your pencils, people

Related

What if America had a “Media Wallet” by Yoni Greenbaum. When I worked with Piano, I saw why this kind of model struggled. The idea was clean. The behavior wasn’t. People didn’t want another password. They didn’t want to stop and think before they could read. The problem wasn’t the price. It was friction.

This page made FTAV’s further reading on the Financial Times for Friday, February 20, 2026.

Bonus links

What’s next for Chinese open-source AI by Caiwei Chen. The adoption of Chinese models is picking up in Silicon Valley, too. Martin Casado, a general partner at Andreessen Horowitz, has put a number on it: Among startups pitching with open-source stacks, there’s about an 80% chance they’re running on Chinese open models…. (related: Please Don’t Say Mean Things about the AI That I Just Invested a Billion Dollars In, generative ai antimoats)

Why the World Is Drawing a Line on Social Media for Kids by Jon Haidt. (As far as I know, teens in Australia can still make GitHub and Wikipedia accounts. How did they manage to slice the definition of “social media”?)

EU Parliament blocks AI features over cyber, privacy fears by Ellen O’Regan and Max Griera. The latest move to switch off AI tools concerns built-in features like writing and summarizing assistants, enhanced virtual assistants and webpage summaries in both tablets and phones, an EU official said, granted anonymity to disclose details of the security policy.

Journalism Is Dead. Long Live Journalism by Rebecca Solnit. Silicon Valley created and abets this chaos, both by undermining the financial basis for traditional news by siphoning away its advertising revenue and audiences, and by creating tools and platforms where, over and over, from Facebook to Substack, the bosses insist they are defending free speech by not filtering out dangerous disinformation and hate speech.

EPIC Crafts 2026 Model Bill to Bolster Age-Appropriate Design Code Laws by Austin Jenkins. EPIC, which filed an amicus brief in the California case, said its model bill was carefully designed to avoid First Amendment issues and was built off of Vermont’s law, which was passed last year after input from EPIC staff.

People who use chatbots for news consider them unbiased and “good enough,” new study finds | Nieman Journalism Lab by Andrew Deck. The study found that right now even regular chatbot users aren’t relying on them alone for news. Most interviewees continued to access traditional news sources, while using chatbots to supplement these habits.

(Previously: Alameda Linux Installfest)

Who: People interested in digital freedom on a budget

What: Chromebook conversion session

Where: StrayCap Multispace, 948 C St., Hayward, California

When: Saturday, February 21, 2026 noon-3pm

Why: A laptop that can be given away to people who need one? In this economy?

StrayCap Multispace is a new not-for-profit Makerspace in Hayward. They have hundreds of expired former school system Chromebooks. The hardware is working, but these laptops have passed their “Automatic Update Expiry” (AUE) date, and no longer receive Chrome OS updates.

Peter Mui from Fixit Clinic asks,

• How can we best re-purpose these in our communities to improve digital literacy, digital sovereignty, digital inclusion, digital equity?

• Depending on those use cases: which distro(s) work best?

• How much can we automate the installation and post-installation process?

Attendees will be welcome to take an ex-Chromebook home to experiment with, or more if you need them.

What to bring:

• Your normal installfest kit if you want

• Tool kit (if you have an electric screwdriver with small bits it will help—these things have a lot of screws.)

• USB drives with your favorite Linux distributions (x86-64)

RSVP so we can get a head count (select “2026-02-21 USA-CA-Hawyard ExChromebook Installfest at StrayCap Noon” from the pulldown menu.)

Why I’m optimistic about these

Most of the web sites that people really need to get to were designed when the hardware specs of these laptops were midrange. So these laptops will still work as laptops. LibreOffice hasn’t gotten any more resource-intensive, either. The cheat code to make these things useful, I think, will be to set them up to not run all the extra surveillance code that has built up since they were new.

Back in 2015, Georgios Kontaxis and Monica Chew found a 44% median reduction in page load time and 39% reduction in data usage when turning on a Firefox tracking protection feature—Tracking Protection in Firefox For Privacy and Performance—and the surveillance advertising overhead has gotten worse over the years since then.

With a little tweaking, I’m optimistic that I can double the effective specs of these machines by blocking the surveillance ads. Code that you choose not to run is the most efficient code. As a side effect, people who have a properly set up ad blocker are more engaged with the web, and are more satisfied with the products and services they buy, and happier in general. Yes, “engagement” can sometimes mean “mental rat hole” so I’m going to be sure to set these up with the cleaned-up version of YouTube so people can watch a video someone sends them without getting sucked into whatever The Algorithm wants to suck them into.

Most people aren’t going to be worried about ad blocking as free riding. Even everyone’s favorite subscription-based crossword puzzle site has ad blocker recommendations now. (More: B L O C K in the U S A)

Anyway, hope to see you in Hayward.

Bonus links

California’s Consumer Privacy Act (CCPA) Assists a Private Right of Action-Shah v. MyFitnessPal by Eric Goldman. (Some more good news from the California courts. Top-down privacy regulation isn’t enough. Juries have more common sense about surveilance norms than state bureaucrats do.)

Article: The Fallen Apple by Matt Gemmell. Tim Cook is giving up his reputation in order to preserve not the company or its people, but rather the stock price. Is that noble? I’m sure that it’s seen as such in the context of American capitalism, but Cook at this point is so utterly contaminated and compromised that I think it’ll be as much of a relief to the man himself as to Apple’s employees when he finally bows out after weathering the remainder of this darkest chapter in modern US history. If, indeed, that chapter does come to an end on schedule. (When buying new IT products means supporting the autogolpe, retrocomputing becomes a patriotic duty)

The TIDALWAVE report from The Heritage Foundation is out.

TIDALWAVE is a progressive Artificial Intelligence–enabled model and computer simulation to simulate a protracted conflict between the U.S. and the People’s Republic of China (PRC). TIDALWAVE identifies gaps and deficiencies and corresponding solutions to resolve anticipated shortfalls in our ability to project and sustain the joint force and to exploit adversary vulnerabilities in order to deplete their ability to conduct military operations.

If we do get into a war, we run out of missiles and torpedoes, real fast.

U.S. munitions inventories, particularly for critical long-range precision-guided munitions (LR-PGMs) and heavyweight torpedoes, are almost certainly insufficient to meet the consumption rates projected for any high-intensity conflict.

Meanwhile, the PRC is rapidly growing manufacturing, but has to import more and more of their oil. So they’re building the biggest, cheapest electric grid they can, as fast as possible (using coal, solar, whatever) which will leave them with more fuel for ships and aircraft. The relative position of the two sides is shifting toward the PRC.

And the report leaves out an important vulnerability for the USA. TIDALWAVE was done using the big, famous LLMs, which might help explain that. A real weak point of LLMs seems to be in generating output about relationships between facts that are connected in some way, but that people talk about using different terminology. And the connection between privacy issues and national security is fundamental but, because defense writers and privacy writers use different words, the LLM is going to miss something.

The manufacturing and cargo handling workers responsible for making all that high-tech ordnance and getting it to where it’s needed are not numerous, and each person has a lot of hard-to-replace know-how. And usually each worker spends a significant fraction of their day in a vehicle with a mobile phone and a license plate.

An adversary can assemble cheap, lethal drones inside the USA, and target the key defense workers, on their commute or at home, using commercially available products and services, from location tracking to SIM cards. Or, if the adversary has a few agents working at certain companies, they get the targeting data free with no paper trail. Some adversary countries are probably trying both ways.

The report points out that Hydroxyl-Terminated Polybutadiene (HTPB) is a critical material for the rocket motors used in missiles. If a defense plant were able to hire a replacement for a manufacturing worker, would they know what to do with a shipment of HTPB if they were able to get one? Raytheon needed to bring back retirees to teach new workers how to build a Stinger missile. But instead of protecting key personnel from targeting, we have a whole industry based on collecting, selling, and sharing personal data. Technically we have a Protecting Americans’ Data from Foreign Adversaries Act (PADFA) and the FTC has been warning Data Brokers about complying with it, but it’s nowhere near enough.

• Hiding company ownership is feasible. We don’t know which businesses here are really “entities controlled by a foreign adversary country.”

• Employees and service providers of a data broker may be foreign adversary controlled even if, on paper, the data broker company is not. (How many employees of the Big Tech companies are non-official cover agents of the PRC?)

A drone raid (think Operation Spiderweb, but targeting individuals using mobile tracking data instead of targeting aircraft using airbase maps, and with more advanced drones) would be relatively easy to pull off in an unprepared USA on the first day of the war, and—for a while anyway—the attacker could keep it deniable. Yes, in the long run, historians will figure out who was responsible, but in today’s media environment the USA could be tricked into blaming someone else for long enough to confuse everybody.

• Throw some money and realistic support at actual terrorist groups that would loudly claim responsibility. Rely on elected officials in the USA to brag on success and amplify the story.

• Run a campaign of good fake videos and influencer content for fake stories.

• Release bad, obvious AI, fakes similar to whatever real evidence gets out, rely on genuine well-intentioned debunkers to create and amplify the narrative that the real attack was fake.

What to do about it

Privacy laws, regulations, and private right of action are part of the survivability onion for defense assets. It is more important than ever to address the tension between the role claimed by US-based tech companies as “national champion” exporters, their roles as enablers of internal violence and data collection, and their role in adding to national security concerns. The same data processing that makes Big Tech so politically popular with the current administration, but less and less popular otherwise, is adding to national security risks that are harder and harder to address.

An expanded Daniel’s Law that covers military personnel and defense workers is going to be one step along the way. We also need more research and testing. The good news is that privacy tech is easier to adopt in a distributed way than other defensive measures. US Navy exercises successfully “destroyed” the fleet at Pearl Harbor in 1932 and 1938, but the knowledge didn’t prevent the real attack a few years later. Defending from a surveillance-enabled attack on defense workers is different, though, because the attack would rely on the same enabling technology as ordinary privacy violations. A strong private right of action for individuals, and a private right of action for employers to protect key personnel, would enable more incremental, case by case, upgrades to personal/national defense—without waiting for one large attack.

Bonus link

This Is Not Sparta by Sarah E. Bond.