Start Up No.1559: Citizen app’s wild inside rise, cause of vaccine clots found, Texas winter deaths ‘undercount’, Instagram’s unLikes, and more

Hard drives prices have risen 50% in the past few months as the Chia cryptocraze has taken off. Unfortunately. CC-licensed photo by Robert Scoble on Flickr.

You can sign up to receive each day’s Start Up post by email. You’ll need to click a confirmation link, so no spam.

A selection of 10 links for you. Here for you. I’m @charlesarthur on Twitter. Observations and links welcome.

‘FIND THIS FUCK:’ inside Citizen’s dangerous effort to cash in on vigilantism • Vice

Joseph Cox:


Andrew Frame was excited. 

It was Saturday night two weeks ago, and Frame, the CEO of the crime and neighborhood watch app Citizen, was on Slack, whipping himself and his employees into what he’d later call at an all-hands meeting a “fury of passion” about a wildfire that had broken out earlier that afternoon in Los Angeles’ Pacific Palisades neighborhood. 

Citizen had gotten a tip that the wildfire was started by an arsonist, and Frame had decided earlier in the night that the fire was a huge opportunity. Citizen, using a new livestreaming service it had just launched called OnAir, would catch the suspect live on air, with thousands of people watching. Frame decided the Citizen user who provided information that led to the suspect’s arrest would get $10,000. Frame wanted him. Before midnight. As the night wore on, Citizen got more information about the supposed suspect. They obtained a photo of the man, which they kept up on the livestream for large portions of the night. More information trickled in through a tips line Citizen had set up. (Citizen said “The information about the person of interest came from an on-the-ground tip from an LAPD Sergeant, followed by emails from local residents who had been approached by LAPD officers.”)

“first name? What is it?! publish ALL info,” Frame told employees working in a Citizen Slack room who were working on the case. 


“BREAKING NEWS. this guy is the devil. get him,” Frame said. “by midnight!@#! we hate this guy. GET HIM.”


Slack really is a boon to journalism these days. Screenshot, email, boom. Also: here’s the problem with people who think they know they’re right, and never ask whether they might be wrong.
unique link to this extract

Industry groups sue to stop Florida’s new social media law • The Verge

Adi Robertson:


Two tech industry organizations have sued Florida over its newly passed rules for social networks. NetChoice and the CCIA — which represent Amazon, Google, Intel, Samsung, Facebook, and other tech giants — say SB 7072 violates private companies’ constitutional rights. They’re asking a court to prevent the law from taking effect, calling it a “frontal assault on the First Amendment.”

SB 7072, which Florida Governor Ron DeSantis signed earlier this week, restricts how large social apps and websites can moderate user-generated content. It makes banning any Florida political candidate or “journalistic enterprise” unlawful, lets users sue if they believe they were banned without sufficient reason, requires an option to “opt out” of sorting algorithms, and places companies that break the law on an “antitrust violator blacklist” that bars them from doing business with public entities in Florida. Notably, it includes an exception for companies that operate a theme park.

NetChoice and the CCIA say SB 7072 conflicts with both constitutional protections and federal Section 230 rules.


Didn’t take long.
unique link to this extract

Bitcoin rival Chia ‘destroyed’ hard disc supply chains, says its boss • New Scientist

Matthew Sparkes:


Chia, a cryptocurrency intended to be a “green” alternative to bitcoin has instead caused a global shortage of hard discs. Gene Hoffman, the president of Chia Network, the company behind the currency, admits that “we’ve kind of destroyed the short-term supply chain”, but he denies it will become an environmental drain.

Bitcoin requires so-called miners to do vast amounts of useless calculations to maintain the network, a system that is known as proof of work. The most recent studies show that bitcoin may currently consume 0.53% of the world’s electricity supply. Chia instead uses a proof-of-space approach that ditches these calculations and relies on empty hard disc space. The more space a miner devotes to the task, the higher their probability of receiving new coins.

In theory, this would consume less energy, but there has been a surge in demand for hard discs since the currency launched earlier this year. Around 12 million terabytes of hard disc space is currently being used to mine Chia, having risen on an exponential curve since its launch in March. When New Scientist first reported on Chia just two weeks ago, that figure was only at 3 million terabytes.

These discs still require energy to produce and run, and there are reports that the constant reading and writing involved in mining can wear them out in weeks, rendering them useless. Hoffman says this problem only affects the cheapest discs.


Well that escalated quickly. Prices for hard drives have risen by 50% at suppliers. To think there was a time when we thought email spam was a problem.
unique link to this extract

Texas’s winter storm death toll likely four times higher than reported • Buzzfeed News

Peter Aldhous, Stephanie Lee and Zahra Hirji:


The true number of people killed by the disastrous winter storm and power outages that devastated Texas in February is likely four or five times what the state has acknowledged so far. A BuzzFeed News data analysis reveals the hidden scale of a catastrophe that trapped millions of people in freezing darkness, cut off access to running water, and overwhelmed emergency services for days.

The state’s tally currently stands at 151 deaths. But by looking at how many more people died during and immediately after the storm than would have been expected — an established method that has been used to count the full toll of other disasters — we estimate that 700 people were killed by the storm during the week with the worst power outages. This astonishing toll exposes the full consequence of officials’ neglect in preventing the power grid’s collapse despite repeated warnings of its vulnerability to cold weather, as well as the state’s failure to reckon with the magnitude of the crisis that followed.

Many of the uncounted victims of the storm and power outages were already medically vulnerable — with chronic conditions including cardiovascular disease, diabetes, and kidney problems. But without the intense cold and stress they experienced during the crisis, many of these people could still be alive today.


unique link to this extract

“Vaccine-induced Covid-19 mimicry” syndrome: splice reactions within the SARS-CoV-2 Spike open reading frame result in Spike protein variants that may cause thromboembolic events in patients immunized with vector-based vaccines • Research Square

Eric Kowarz and others at Goethe University, Frankfurt:


Soluble Spike variants together with newly built antibodies against Spike protein as well as the highly specific blood flow conditions in the central venous sinus of the brain may result in the rare but severe events after vaccination observed with ADZ1222/Vaxzevria. Noteworthy, the vaccine from Johnson & Johnson appears to carry fewer splice donor sequences, especially SD506 and SD3614 (see Table 1B and 1C), which are the strongest predicted splice donor sites in the AZD1222 sequence (see Fig. 1A). This may explain the ~ 10-fold lower incidence of severe side effects with the Johnson & Johnson vaccine when compared to the AZD1222 vaccine.

In principle, such thromboses may occur in any site of the human body where endothelial cells express ACE2. Soluble Spike proteins which still exhibit the important core portion of the S1 domain (R319-F551) will be able to bind these receptors. When the immune system starts to produce antibodies against the Spike protein, the endothelial cells will not only bind the soluble Spike protein variants, but would also be decorated with the newly formed antibodies. This will give rise to strong inflammatory reactions either by ADCC (antibody dependent cell-mediated cytotoxicity) or CDC (complement dependent cytotoxicity) occurring in these vessels at various sites where such soluble Spike protein variants accumulate.


OK, so this is very technical, but it’s the explanation for why the non-mRNA vaccines (eg AstraZeneca, less so Johnson & Johnson) can occasionally provoke blood clots, particularly in the brain. It also explains what needs to be done to lessen that risk. If confirmed – it’s still a preprint – a very important paper.
unique link to this extract

What Instagram really learned from hiding like counts • The Verge

Casey Newton:


After more than two years of testing, today Instagram announced what it found: removing likes doesn’t seem to meaningfully depressurize Instagram, for young people or anyone else, and so likes will remain publicly viewable by default. But all users will now get the ability to switch them off if they like, either for their whole feed or on a per-post basis.

“What we heard from people and experts was that not seeing like counts was beneficial for some, and annoying to others, particularly because people use like counts to get a sense for what’s trending or popular, so we’re giving you the choice,” the company said in a blog post.

At first blush, this move feels like a remarkable anticlimax. The company invested more than two years in testing these changes, with Mosseri himself telling Wired he spent “a lot of time on this personally” as the company began the project. For a moment, it seemed as if Instagram might be on the verge of a fundamental transformation — away from an influencer-driven social media reality show toward something more intimate and humane.

In 2019, this no-public-metrics, friends-first approach had been perfected by Instagram’s forever rival, Snapchat. And the idea of stripping out likes, view counts, followers and other popularity scoreboards gained traction in some circles — the artist Ben Grosser’s Demetricator project made a series of tools that implemented the idea via browser extensions, to positive reviews.

So what happened at Instagram?

“It turned out that it didn’t actually change nearly as much about … how people felt, or how much they used the experience as we thought it would,” Mosseri said in a briefing with reporters this week. “But it did end up being pretty polarizing. Some people really liked it, and some people really didn’t.”


Very meta, that people liked or didn’t like when there were or weren’t Likes.
unique link to this extract

Facebook no longer treating ‘man-made’ Covid as a crackpot idea • POLITICO

Cristiano Lima:


Facebook will no longer take down posts claiming that Covid-19 was man-made or manufactured, a company spokesperson told POLITICO on Wednesday, a move that acknowledges the renewed debate about the virus’ origins.

Facebook’s policy tweak arrives as support surges in Washington for a fuller investigation into the origins of Covid-19 after the Wall Street Journal reported that three scientists at the Wuhan Institute of Virology were hospitalized in late 2019 with symptoms consistent with the virus. The findings have reinvigorated the debate about the so-called Wuhan lab-leak theory, once dismissed as a fringe conspiracy theory.

President Joe Biden said Wednesday that he has ordered the intelligence community to “redouble” its efforts to find out the virus’ origin and report back in 90 days. Biden also revealed that the intelligence community is split between two theories about Covid-19’s origin, and said the review will examine “whether it emerged from human contact with an infected animal or from a laboratory accident.” Bipartisan support is also building on Capitol Hill for a congressional inquiry.

But the focus of late has been on the notion that the virus may have accidentally escaped from the lab, not that it was man-made or purposely released — theories that could now propagate on Facebook. Genetic studies of the virus have found flaws in the protein it uses to bind to human cells. Those are features that someone trying to engineer a bioweapon likely would have avoided.

Shifting definitions on social media: Facebook announced in February it had expanded the list of misleading health claims that it would remove from its platforms to include those asserting that “COVID-19 is man-made or manufactured.”


This is slightly exhausting. I didn’t link earlier this week to the WSJ report (three people from the Wuhan Institute of Virology went to hospital, like everyone in China who’s a bit ill; that’s pretty much it) because of a comprehensive demolition on Twitter by an account I trust of the story’s lack of rigour, and obvious non-intelligence-community derivation.

But it’s been taken up as being somehow “evidence” that the lab leak hypothesis is more likely than it was a week ago. That’s not true. It’s still absurd to claim that SARS-Cov-2 is man-made or manufactured: that’s a fringe conspiracy theory. It’s possible to think there could have been an accidental leak of something, but there’s still zero evidence. A Congressional inquiry will discover absolutely nothing, except there’s lot of non-scientists who like the sound of their own prejudices.
unique link to this extract

Google medical researchers humbled when AI screening tool falls short in real-life testing • TechCrunch

Devin Coldewey, writing in April 2020:


The Google system was intended to provide ophthalmologist-like expertise in seconds. In internal tests it identified degrees of DR with 90% accuracy; the nurses could then make a preliminary recommendation for referral or further testing in a minute instead of a month (automatic decisions were ground truth checked by an ophthalmologist within a week). Sounds great — in theory.

But that theory fell apart as soon as the study authors hit the ground. As the study describes it:


We observed a high degree of variation in the eye-screening process across the 11 clinics in our study. The processes of capturing and grading images were consistent across clinics, but nurses had a large degree of autonomy on how they organized the screening workflow, and different resources were available at each clinic.

The setting and locations where eye screenings took place were also highly varied across clinics. Only two clinics had a dedicated screening room that could be darkened to ensure patients’ pupils were large enough to take a high-quality fundus photo.


The variety of conditions and processes resulted in images being sent to the server not being up to the algorithm’s high standards:


The deep learning system has stringent guidelines regarding the images it will assess…If an image has a bit of blur or a dark area, for instance, the system will reject it, even if it could make a strong prediction. The system’s high standards for image quality is at odds with the consistency and quality of images that the nurses were routinely capturing under the constraints of the clinic, and this mismatch caused frustration and added work.


Images with obvious DR but poor quality would be refused by the system, complicating and extending the process. And that’s when they could get them uploaded to the system in the first place.


Since I was wondering yesterday about Google’s deal with a US hospital, and how medical data had gone. The answer is: not always well. And then there’s the data….
unique link to this extract

A software error made Spain’s child Covid mortality rate seem to skyrocket • Slate

Elena Debré:


On March 10, a respected peer-reviewed medical journal, the Lancet, published Spain’s child Covid mortality rate as around two to four times that of the U.S., U.K., Italy, Germany, France, and South Korea. The paper said that 54 children (defined as below 19) had died of Covid in the small country, making Spain’s reported death rates a staggering 4.9% for kids aged 10-19—which is at least 2.92 percentage points higher than other country in the report.

Right after the Lancet paper was published, Pere Soler, a pediatrician at hospital in Catalonia, started getting calls. Concerned reporters were trying to reach him for comment. “The first question that I received was, ‘Have you been lying to us?'” Soler says. He and other prominent pediatricians around the country had been in close contact with a circle of reporters throughout the pandemic, keeping them updated on child Covid research and school reopenings. This high of a child mortality rate did not add up with the numbers doctors had been seeing and feeding to the media.

As a re-examination of the information would soon reveal, in reality, only seven children had died of Covid. (The Lancet data has since been corrected.)

“Even though I didn’t know what the problem was, I knew it wasn’t the right data,” Soler realized once he got his hands on the Lancet paper. “Our data is not worse than other countries. I would say it is even better,” he says.


Turned out the system they used couldn’t handle people aged over 100: it only expected people to live to 99.
unique link to this extract

Nine things we learned from the Epic v. Apple trial • The Verge

Russell Brandom:


It’s been just over three weeks since the Epic v. Apple proceedings kicked off, and the news has been relentless. So as we wait for a verdict to roll in, we’re taking a quick turn through all the biggest takeaways from the trial. A lot of the juiciest points didn’t speak directly to the verdict — like the profit structure of the Xbox or the troubled history of Fortnite crossplay — but that’s part of the fun of a massive trans-corporate dustup like this. Once you start digging through CEO Tim Cook’s inbox, all sorts of interesting stuff comes out.


It’s not just Apple stuff: there’s also Epic Games’s weird economics of its store, Xbox’s hardware economics, and how Apple slow-walks progressive web apps.
unique link to this extract

Errata, corrigenda and ai no corrida: none notified

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.