If you read this newsletter with any regularity, you know I often write about Facebook. Last month, the WSJ released what it called the Facebook Files, stories based on a trove of documents leaked by Frances Haugen, a former Facebook employee and whistleblower. Then, earlier this month Haugen invited journalists from 17 other media outlets (the number has since grown) to a call and said she’d turn over the documents to them so they could all report different pieces of what they found. So, for the last week, a lot of stories have come out about Facebook. Here is a handy Google Doc tracking all the stories about Facebook, which I think you will agree is a lot of stories.
There are a few insights to be gleaned from the information tsunami released over the last week, however. Facebook has maintained a culture of extreme secrecy for years, claiming all its decisions and technology were trade secrets, and intimating they had reasons for doing what they did, often claiming they had the public’s best interests in mind.
This was, to put it mildly, a weak argument in the face of the many scandals exposing Facebook’s corporate malfeasance. One thing the Facebook Files have revealed is how arbitrary many of the platform changes were. Here is a story from the Washington Post about Facebook’s scoring system for reactions and comments on newsfeed posts:
Starting in 2017, Facebook’s ranking algorithm treated emoji reactions as five times more valuable than “likes,” internal documents reveal. The theory was simple: Posts that prompted lots of reaction emoji tended to keep users more engaged, and keeping users engaged was the key to Facebook’s business.
[…]
The warning proved prescient. The company’s data scientists confirmed in 2019 that posts that sparked angry reaction emoji were disproportionately likely to include misinformation, toxicity and low-quality news.
For three years, this one change to Facebook’s algorithm - perhaps made with good intentions - dramatically affected the kind of content people saw on the platform. We talk a lot here about algorithms, and here is how Facebook’s works:
Facebook takes into account numerous factors — some of which are weighted to count a lot, some of which count a little and some of which count as negative — that add up to a single score that the news feed algorithm generates for each post in each user’s feed, each time they refresh it. That score is in turn used to sort the posts, deciding which ones appear at the top and which appear so far down that you’ll probably never see them. That single all-encompassing scoring system is used to categorize and sort vast swaths of human interaction in nearly every country of the world and in more than 100 languages.
Essentially, Facebook assigned different values to different things, totaled them up into a “score” and that determined how likely people were to see a post organically - not served as an advertisement - in their news feed. When they rolled out the emoji reactions in 2017, it was an attempt to reverse a decline in people posting on the platform. Giving more weight to the new emojis created a feedback loop - posts with emoji reactions started showing up more in peoples’ feeds, which meant they used more reactions, driving up the scores of the posts with the most interactions.
The problem was, Facebook gave the angry emoji the same 5x score as a heart or a sad emoji, which meant content that pissed people off was suddenly flooding their feeds, as they angry-faced it.
A year after the emojis launched, Facebook downgraded the angry emoji…by one multiple. So it was still four times more potent than a like. By then, the company had evidence the angry emoji was directly linked to misinfo and toxic content, but they resisted calls to downgrade the emojis their researchers told them amplified bad posts on the platform.
Replies to posts were even more potent, weighted thirty times a like, which elevated content that had arguments in the replies, or, in some cases, allowed unscrupulous marketers to boost their content, which Facebook finally took action against…in 2020:
It was part of a broader fine-tuning of signals. For example, single-character comments would no longer count. Until that change was made, a comment just saying “yes” or “.” — tactics often used to game the system and appear higher in the news feed — had counted as 15 times the value of a like.
This story is told over and over again - Facebook had the same issues rolling out live video, then Reels, then Facebook Stories, etc - and it goes something like this:
Facebook is concerned engagement numbers are declining
Facebook rolls out a new feature to improve engagement
Facebook’s engineers heavily weight the use of this new feature
Spammers, scammers, and disinfo merchants start using the new feature
The quality of content on Facebook declines
Facebook is concerned engagement numbers are declining
If only there were some way to fix this problem! Like, I don’t know, if they’d listened to their researchers who were apparently literally yelling in company chatrooms about ways to clean up the news feed, show posts that made people feel better rather than worse, and prevent scammy advertisers from taking over the platform.
One problem with running a public company is you are encouraged to think in three month increments, because that’s how often you report your earnings. Facebook could be immune to this sort of short-termism if it wanted to, because Facebook makes obscene amounts of money and no one is going to stop buying Facebook stock if they have a quarter where they only make, like, three billion dollars instead of four. They could use this leeway to invest in cleaning up their platform, or increasing transparency about how their systems work, or simply pulling out of markets they are unable to safely operate in - think India, Myanmar, etc. But that’s never been how Facebook operates, because Mark Zuckerberg - who has effective total control over the company through his supervoting shares - thinks the most important thing for Facebook is to grow as big as possible, forever, and to constantly roll out new features and enter new markets to achieve that goal.
I don’t know what is going to happen to Facebook after the storm dies down, maybe people at the company will do some real reflection on what they’ve created and how it can be fixed, or maybe they’ll just hire more lawyers and PR companies and forge ahead because they’ve got too much money and even the harshest forms of penalties - criminal investigations, new laws, heavy regulation - would take years to wind through the courts. Perhaps the only bright spot this week is news that tech employees are starting to turn down job offers at Facebook, which may give them pause - without a significant churn of tech workers, Facebook won’t be able to expand endlessly and create new tools with which to make all our lives just a bit more miserable.
Meta
Yesterday Facebook announced it was rebranding itself to Meta. Here is Ed Zitron picking apart why, tone deaf timing aside, there really isn’t much of anything to Meta, other than some nice-sounding words:
Facebook has and will continue to be the place where billions of people connect, and I believe its social side will continue to do something, though its user growth is slowing. What should trouble you is that Meta/Oculus has only sold a few million headsets total. This would be fine - hell, it would be positive - if Zuckerberg hadn’t just shifted his entire company to focus on the metaverse, with the primary way of interacting with said metaverse being a headset that even in its best iteration still has to work out how to stop giving people motion sickness.
I have spoken to vision researchers about this very problem, and there is a large percentage of the population who get sick in a VR headsets, and even Facebook’s most optimistic estimates say their tweaks are a couple years away.
And this, in a nutshell, is the problem. If Zuck had left Oculus alone for a couple more years and they miraculous produced VR and AR headsets everyone could wear and not puke - unlikely! - he could have credibly rebranded the company then, instead he’s rolling it out in the midst of a dozen scandals and a PR firestorm and it makes no sense:
The technology required to get even a few steps into the totally fictional metaverse world Mark has rendered in his agonizing 11-minute long video is so far off that it is honestly irresponsible for a public company to act as if it’s possible. And even when it is, how much will it cost? How easily will it be available? What about poorer countries that have enjoyed using services like WhatsApp and Facebook for free - are they left out of the metaverse?
It is entirely possible that this is simply another misstep, and Facebook will realize Meta makes no sense and walk it back, or say hey, just kidding, it’s actually just selling another million Oculus headsets, or who knows. Maybe it’s a legal ruse to distance the company from the problems with Facebook’s social platform. What it is not is a real product, or a realistic vision for Fa- sorry, Meta in anything resembling the near term, and hopefully the media ask the tough questions necessary to expose the sham.
TMTG
Speaking of companies without a coherent business model, let’s talk about Trump’s media venture! Last week we talked a little about it, and now we have…more information, though not very much:
Former President Donald Trump is gearing up to unveil his new social media venture, saying the project will “challenge the dominance of the Big Tech giants” and will address the “corruption” of major tech platforms.
Right, okay, sure. Matt Levine points out the only bit of news in this news release:
“Last week,” Trump says, “I announced the creation of a major new company.” Ordinarily, there are two distinct important events in the life of a SPAC company, generally quite separate in time: One day the company is created, and then, probably years later, it announces a SPAC merger.
There is Trump admitting he created TMTG and then immediately merged it with DWAC via a SPAC - I am dying typing this - and suddenly the company had $293 million dollars, which DWAC had sitting around waiting for a company to take public. Normally, you’d start a company and run it for awhile, and once you had a viable product, or a business plan, or a need for investment money, you’d find a SPAC and convince them to give you money, but if you’re Trump you cut the deal in advance, using an imaginary company with no business plan or whatever, and congratulations, you’ve got a multibillion dollar company. Finance!
How much is Trump poised to make off this? Potentially quite a bit, if the stock price stays where it is now:
The basic deal is that Trump and friends get about 86 million shares of the combined company, while DWAC’s public shareholders get about 29 million and its sponsors get another 7 million. The earnout gives Trump and friends an extra 40 million shares if the stock is above $30 consistently after the deal closes. Assuming that it will be — it has never closed below $45 since the deal was announced — Trump will have about 126 million shares worth, at yesterday’s closing price of $59.07, about $7.4 billion.
DWAC is at $71 at the time I’m writing this, so Trump stands to make even more, though his shares will be locked up for awhile. There is something very funny about Trump spending his whole adult life pretending he was a billionaire when he wasn’t, spending four years trying to steal as much money as he could from the US government and mostly losing money, and now finally joining the three comma club by founding a phony media company.
Trump likely left billions of dollars on the table when he took his fake company public:
The press release said the new company would have an initial enterprise value of $875 million. Assuming that calculation used a $10 a share offering price and the $293 million Digital World has in trust, the SPAC’s owners will get about 42% of the combined company after accounting for shares the sponsor receives if a deal gets done.
At current valuations, he sold over $3 billion worth of stock for $293 million dollars, which is about what you’d expect from the master of The Deal.
Nursing Homes
One byproduct of modern medicine is that humans are living a lot longer than they used to, and the assisted living and nursing care businesses are booming. Caring for aging seniors with a variety of chronic health conditions is not an easy job on a good day, but the nursing care system in the US has come under repeated scrutiny for poor quality of care, as much of it is - of course! - for-profit, creating the same perverse incentives for care facilities we’ve previously talked about at hospitals.
Many nursing patients have some form of dementia, which makes them more difficult for staff to care for. One thing nursing homes have done for years is prescribing antipsychotic drugs to keep them under control. Problem is, prescribing these drugs to elderly patients with dementia is wildly dangerous:
Antipsychotic drugs — which for decades have faced criticism as “chemical straitjackets” — are dangerous for older people with dementia, nearly doubling their chance of death from heart problems, infections, falls and other ailments. But understaffed nursing homes have often used the sedatives so they don’t have to hire more staff to handle residents.
The government has raised concerns about overmedication of seniors in homes since the seventies, and these drugs double their chance of death, but nursing homes have leaned heavily on antipsychotics to save money on staff. Excellent.
The drugs are so dangerous that the government requires homes to report the number of patients currently taking them. Good, right? Well:
But there is an important caveat: The government doesn’t publicly divulge the use of antipsychotics given to residents with schizophrenia or two other conditions.
But, surely the homes aren’t misdiagnosing patients with schizophrenia to administer deadly drugs! Yeah, well:
The share of residents with a schizophrenia diagnosis has soared 70 percent since 2012, according to an analysis of Medicare data. That was the year the federal government, concerned with the overuse of antipsychotic drugs, began publicly disclosing such prescriptions by individual nursing homes.
Today, one in nine residents has received a schizophrenia diagnosis. In the general population, the disorder, which has strong genetic roots, afflicts roughly one in 150 people.
Schizophrenia, which often causes delusions, hallucinations and dampened emotions, is almost always diagnosed before the age of 40.
The incentives for the nursing homes are twofold - if they prescribe high rates of antipsychotics, they can take a hit to their “star” rating from the government. If they hire more staff, they might have to pay them, which means they would make less money, so that’s obviously out. So, instead, they’re having doctors declare them suddenly schizophrenic. Piece of cake!
How widespread is the problem?
The figures [obtained by the Times] showed that at least 21 percent of nursing home residents — about 225,000 people — are on antipsychotics.
Twenty one percent of patients in nursing homes are on drugs that are likely to kill them, or at the very least make them susceptible to other serious ailments. Not to mention, by putting people in “chemical straitjackets” they’re zombifying them for the remaining time they have alive. The stories in the article are horrifying, reports of people in a near permanent state of half-consciousness.
As the government has attempted to curtail the over-prescription of antipsychotics in recent years, homes have turned to epilepsy drug Depakote to keep their dementia patients subdued:
Between 2015 and 2018, the most recent data available, the use of anti-seizure drugs rose 15 percent in nursing home residents with dementia, according to an analysis of Medicare insurance claims that researchers at the University of Michigan prepared for The Times.
And while Depakote’s use rose, antipsychotic prescriptions fell 16 percent.
This drug is not for treating dementia, it is for making old people drowsy. Workers at nursing homes may even be…hiding it in residents’ food?
About half the complaints that California Advocates for Nursing Home Reform receives about inappropriate drugging of residents involve Depakote, said Anthony Chicotel, the group’s top lawyer. It comes in a “sprinkle” form that makes it easy to slip into food undetected.
“It’s a drug that’s tailor-made to chemically restrain residents without anybody knowing,” he said.
Oh, and lest we put the blame squarely on unethical nursing home doctors and workers, the drug company who makes Depakote was encouraging this behavior in the 2000s:
In the early 2000s, Depakote’s manufacturer, Abbott Laboratories, began falsely pitching the drug to nursing homes as a way to sidestep the 1987 law prohibiting facilities from using drugs as “chemical restraints,” according to a federal whistle-blower lawsuit filed by a former Abbott saleswoman.
According to the lawsuit, Abbott’s representatives told pharmacists and nurses that Depakote would “fly under the radar screen” of federal regulations.
Abbott settled the lawsuit in 2012, agreeing to pay the government $1.5 billion to resolve allegations that it had improperly marketed the drugs, including to nursing homes.
And, last but not least, the for-profit nursing industry’s lobbying group has spent the last year trying to push Congress to expand the list of conditions they can treat with antipsychotics:
Last year, though, the industry teamed up with drug companies and others to push Congress and federal regulators to broaden the list of conditions under which antipsychotics don’t need to be publicly disclosed.
“There is specific and compelling evidence that psychotropics are underutilized in treating dementia and it is time for C.M.S. to re-evaluate its regulations,” wrote Jim Scott, the chairman of the Alliance for Aging Research, which is coordinating the campaign.
Gaaaaahhhhhh. The US leans on nursing homes to treat our elderly because our social safety net is nonexistent, caring for family is either poorly paid or not paid at all, and because dealing with aging relatives who may not be in full command of their senses is a real challenge. The places they end up, however, may be a whole lot worse than the alternative.
MLMs
Here is a depressing story about a bunch of people getting COVID-19 from an MLM conference in Las Vegas:
Celebrate Paparazzi 2021 has been described as a superspreader event. Dozens of the company’s consultants and loved ones were sickened by Covid-19 after the event, sources told me, and at least six people died. While it’s impossible to know exactly where attendees contracted the disease, many became ill after attending the large, often-unmasked convention.
MLMs had a very good year and a half, with people stuck at home and turning to ripping one another off on Facebook as a source of income. Unfortunately, when you join an MLM, in-person networking is a big part of the equation, and a bunch of MLMers apparently breathed virus all over one another, and some of them died.
MLMs have a history of taking advantage of dire moments in US history, because of course they do:
MLMs have historically been very good at seizing on specific moments and trends to recruit sellers and expand their ranks. During the 1960s, Amway made bomb shelters; in the Great Recession, Donald Trump made millions off a multilevel marketing company called ACN, which markets internet, gas, phone, and electricity products.
Revenues - which flow to the ownership and people at the very top, because they are pyramid schemes - are good:
A majority of respondents have consistently reported that the virus has had a positive impact on both US and global revenue, with 60 percent saying so as of late July. The DSA estimates that the industry grew by 13.9 percent in 2020 to $40.1 billion in retail sales.
Everything sucks, man. Anyhow, the FTC might change the rules and force MLM companies to disclose more information to potential recruits. It’s not the ideal solution - which would be to ban them entirely - but the industry is long overdue for increased scrutiny. A 2017 report found that 99% of MLM members lose money, which the MLM industry insists is a legitimate business model.
Short Cons
Slate - “Celebrities are lending their names and promotional talents to everything from NFTs to crypto exchanges to their own custom coin offerings. Together they are communicating something clear but terribly misleading: that cryptocurrencies, and the many shady areas of this gray-market economy, offer a path to sustainable riches.”
WaPo - “A woman is seeking $5 million from Kellogg, accusing the popular cereal company of misleading customers into thinking its strawberry Pop-Tarts contain more strawberries than they actually do.”
Wired - “When German Police took down the dark web marketplace DarkMarket in January, they did more than shutter the largest underground online drug bazaar operating at that time. They also nabbed more than 20 servers in Moldova and Ukraine, infrastructure that contained a host of information about the site’s buyers and sellers.”
Defector - “’It’s gotten harder to steal stuff,’ lamented on stage Howard Marks, the Oaktree Capital founder who at Citigroup in the late 1970s had been one of the first buyers of Milken’s revolutionary debt.”
Tips, thoughts, or metas to scammerdarkly@gmail.com