Start Up No.1801: Facebook’s fear of the right wing, Christchurch’s missed chance, speeding’s real death toll, war by ratio, and more

The Alqueva reservoir is western Europe’s largest artificial lake, and the site of Europe’s largest floating solar farm: 7.5GWh annual generation. CC-licensed photo by Paulo Valdivieso 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 9 links for you. All different. I’m @charlesarthur on Twitter. Observations and links welcome.

Facebook killed News Feed fix for fear of conservative backlash • Gizmodo

Dell Cameron, Shoshana Wodinsky and Mack DeGeurin:


Today, Gizmodo is publishing our third batch of the Facebook Papers—documents that, among other things, shine a light on the company’s reluctance to take action against known sources of misinformation. Employees whose work appears in the papers repeatedly attribute decisions like the shelved News Feed update to fears that the company would be portrayed as favoring certain publications that, in some cases, its own users judged more informative. In particular, accusations of liberal bias by Republican leaders weighed heavily in debates over whether to improve News Feed or correct flaws in the ways Facebook prioritized journalism and other political content. Accusations of a liberal slant controlling the company are described in two papers as playing a crucial role in decisions reached during employee-led efforts to minimize the frequency of propaganda and misinformation in people’s feeds.

An internal post dated August 2019 briefly describes the decision by Facebook to kill a News Feed update purportedly designed to prioritize “high quality” news. In this case, Facebook obtained the underlying data responsible for gauging the trustworthiness of news sources by polling users. The company came to the decision not to reduce the flow of “low quality” news to stave off charges from “some quarters” about “perceived anti-conservative bias,” according to the post.

Asked about the discrepancy between the company’s prior claims and the once-confidential testimony of its own employees, a Facebook spokesperson declined to comment.

In the same 2019 document, Facebook employees estimated that the company had only taken action against “approximately 2% of the hate speech on the platform,” while concluding that misinformation, when noticed at all, often goes unidentified “until after it has gotten a lot of distribution.” The most “impactful abusive accounts,” it says, continue to persistently evade moderation. While employees generally have “considerable leeway” when it comes to making decisions that affect “a wide range of content,” the author writes, “policy concerns become significantly higher” when politics enter the frame.

The documents take on new relevance in the political climate of 2022. Attempts by social media companies to minimize the spread of election-related hoaxes and false news have spurred Republican leaders in several states to pursue new laws around content moderation.


unique link to this extract

Buffalo might never have happened if online hate had been tackled after Christchurch • The Guardian

Imran Ahmed is chief executive of the Center for Countering Digital Hate:


In a joint statement in 2019, Meta, Twitter and Google committed to uphold the Christchurch Call to eliminate terrorist and violent extremist content online. They stated that they would be “resolute in [their] commitment to ensure [they] are doing all [they] can to fight the hatred and extremism that lead to terrorist violence”.

The failure of social media companies to act sufficiently on known racist content connected with terrorism is a violation of their own terms and conditions, the pledges made to an international community when the cameras were rolling, and the dignity that the victims of Buffalo were entitled to have – the right to life.

Social media and online spaces are often where people meet, seek information and become radicalised through a rabbit-hole of lies, hate and misinformation. Those with fringe beliefs will be exposed to increasingly more radical content as a result of recommendation algorithms. The failure of social media giants to effectively tackle online hate and misinformation has real-world impacts. Words can kill.

Perhaps the only thing that explains why – despite so many pledges, so many platitudes and commitments to voluntary frameworks – the social media platforms have failed to act is because of the memo that Andrew Bosworth, now chief technical officer of Meta, wrote to his fellow employees on their internal messaging board, called the Ugly Truth. In it he said: “So we connect more people. That can be bad if they make it negative. Maybe it costs a life by exposing someone to bullies. Maybe someone dies in a terrorist attack co-ordinated on our tools. And still we connect people.”

It is, quite simply, a bald statement of personal indifference to the grief of people, families and our nations.


The Bosworth Memo (it deserves its own capital letters) dates back to June 2016, and always comes across as a distillation – it’s only 420 words – of the sociopathic mentality that can infect people who get disintermediated from those they affect.
unique link to this extract

Speeding causes three times as many road deaths as previously thought • The Sunday Times

Nicholas Hellen:


Under a more accurate reporting system to be adopted by the National Police Chiefs’ Council (NPCC), speed will be noted as a contributing cause of half of Britain’s 1,500 annual road deaths. It is currently noted as the cause of just 375.

At present, official figures record that the biggest cause of deaths on the road is a driver or rider failing to look properly. Loss of control is second, with exceeding the speed limit and driver carelessness joint third.

But the figures, known as STATS19, are based on reports from the scene of the collision, and are not updated even if an investigation finds other factors were involved. Now forces will be asked to record the results of their final assessment of the causes, typically after a forensic review has been conducted.

A Metropolitan Police review has found that speeding was cited as a factor in 17.5% of fatal crashes in 2019 based on the initial investigation, but 49.2% based on the final results. The figure for 2020 was revised up from 19.1% to 46.8%. This included motorists who broke the speed limit and those who drove too fast for the conditions.

Even greater anomalies have been discovered in the Greater Manchester police area, where the official figures for road deaths in the three years from 2016 to 2018 show that motorists were breaking the speed limit in 64% of cases rather than 15%.


That’s quite a big difference. “Too fast for the conditions” isn’t going to be affected by a speed limiter, but the Manchester numbers suggest they could make a difference. Couldn’t they? (Thanks Fabian for the link.)

unique link to this extract

SparkToro & Followerwonk joint Twitter analysis: 19.42% of active accounts are fake or spam • SparkToro

Rand Fishkin:


From May 13-15, 2022, SparkToro and Followerwonk conducted a rigorous, joint analysis of 44,058 public Twitter accounts active in the last 90 days. These accounts were randomly selected, by machine, from a set of 130+ million public, active profiles. Our analysis found that 19.42%, nearly four times Twitter’s Q4 2021 estimate, fit a conservative definition of fake or spam accounts (i.e. our analysis likely undercounts). Details and methodology are provided in the full report below.

For the past three years, SparkToro has operated a free tool for Twitter profiles called Fake Followers. Over the last month, numerous media outlets and other curious parties have used the tool to analyze would-be-Twitter-buyer, Elon Musk’s, followers. On Friday, Mr. Musk tweeted that his acquisition of Twitter was “on hold” due to questions about what% of Twitter’s users are spam or fake accounts.

SparkToro is a tiny team of just three, and Fake Followers is intended for informal, free research (our actual business is audience research software). However, in light of significant public interest, we joined forces with Twitter research tool, Followerwonk (whose owner, Marc Mims, is a longtime friend) to conduct a rigorous analysis answering:

• What is a spam or fake Twitter account?
• What% of active Twitter accounts are spam or fake?
• What% of Mr. Musk’s followers are spam, fake, or inactive?
• Why should our methodology be trusted?
• We address each of these questions below.


First one: what’s a spam/fake account?


“Our definition (which may differ from Twitter’s own) can best be described as follows:

“Spam or Fake Twitter accounts are those that do not regularly have a human being personally composing the content of their tweets, consuming the activity on their timeline, or engaging in the Twitter ecosystem.”


Except that would include entirely beneficial automated accounts that tweet the contents of feeds (such as, oh, @theoverspill) or, like @threadreaderapp, link to concatenated threads – such as the Twitter CEO’s today about how they calculate spam accounts.

Meanwhile, Elon Musk looks more and more like someone casting around for a reason not to buy Twitter.
unique link to this extract

How a simple ratio came to influence military strategy • WSJ

Josh Zumbrun:


Early in Russia’s invasion of Ukraine, it wasn’t just Moscow that believed its offensive could succeed quickly. In February, even U.S. officials warned Kyiv could fall in days.

Russians had numbers on their side, or more precisely a number: the 3:1 rule, the ratio by which attackers must outnumber defenders in order to prevail. It is one of several “force ratios” popular in military strategy. Russia, it seemed, could amass that advantage.

The war in Ukraine has brought renewed interest in force ratios. Other ratios in military doctrine include the numbers needed to defeat unprepared defenders, resist counterinsurgencies or counterattack flanks. Though they sound like rules of thumb for a board game like Risk, the ratios have been taught to generations of both American and Soviet and then Russian tacticians, and provide intuitive support for the idea Ukraine was extremely vulnerable.

“I would imagine that most of them are thinking in those terms, that you need something on the order of a 3:1 advantage to break through,” said John Mearsheimer, a University of Chicago professor whose work focuses on security competition between great powers. “It’s clear in this case that the Russians badly miscalculated.”

…Overall, Russia’s military has quadruple the personnel and infantry vehicles, triple the artillery and tanks, and nearly 10 times the armored personnel carriers, according to the International Institute for Strategic Studies, the London-based think tank.

With 190,000 Russian troops concentrated to invade in February, and Ukraine’s military spread across the country, (only 30,000 troops, for example, were estimated to be in Ukraine’s east near the Donbas region) it appeared Russia had the numbers to overwhelm Ukraine.


Looks like all the textbooks need to be rewritten to cater for drones and directed anti-tank weapons. Though possibly it just needs a section on how not rotating your tyres dilutes your force ratio.
unique link to this extract

New study on SIDS is not what the media says it is • ParentData

Emily Oster:


Knowing that a baby had an enzyme level of 4 [of butyrylcholinesterase, BChE] might indicate a higher risk, but definitely there are many infants with a level of 4 who survive. This is all very distant from being a perfect or close to perfect predictor of anything. 

That’s the main paper finding. What I take from this, with the paper alone, is an interesting possibility that should be explored more. The sample sizes are small, and with a new theory like this we always want to see more out-of-sample testing. It would also be important to understand whether there are other characteristics that differ across these groups that might be related to enzyme levels and could be driving this result. It certainly seems plausible that BChE (and AChE, acetylcholinesterase) levels are a predictor of SIDS risk, and it is worth more follow-up work to understand it. 

My primary frustration is with the media coverage of it and, in particular, the story with the headline I noted above, which it would seem has permeated the Facebook mom groups. There is just nothing in the paper that supports that headline or most of the article. Yes, the paper produces suggestive evidence that there might be a correlation between this enzyme level and SIDS. But to say it “pinpoints” the reason infants die from SIDS is absolutely not true. 

Perhaps more problematic is this paragraph: 


Previously, parents were told SIDS could be prevented if they took proper precautions: laying babies on their backs, not letting them overheat and keeping all toys and blankets out of the crib were a few of the most important preventative steps. 


Though not stated explicitly, the implication is that these steps are now no longer necessary. This is definitely false.


There’s also a long thread by a neurophysiologist saying much the same, though in some more detail.

So it seems as though SIDS is not a solved problem at all.
unique link to this extract

Did Terra operators bail out crypto whales? • New Money Review

Paul Amery:


The operators of the collapsed Terra stablecoin ($UST) last week allowed selected holders of the dollar token to cash out at close to 100 cents in the dollar, using cryptocurrency exchanges Gemini and Binance as a conduit.

According to the Luna Foundation Guard, which operates a reserve pool backing $UST and its related token, $LUNA, holders of $2.7bn in face value of $UST were able to sell them for bitcoin in two transactions last week, one with an effective bitcoin/UST exchange rate of $32,334 and the other with an effective exchange rate of $35,054.

The Luna Foundation Guard revealed this information in a series of tweets released this morning.

It did not disclose the timing of the transactions. However, evidence from Elliptic, a cryptocurrency research firm, suggested that the transactions took place on May 9 and early on May 10, when the $UST price traded in secondary markets as low as 60 cents in the dollar.

…This implies that the holders of $UST who were able to sell their tokens for the bitcoin offered by the Luna Foundation Guard were able to exit their positions at close to face value ($1), rather than the deep discounts on offer in the secondary market.

…The two exchanges are now likely to come under increasing pressure to disclose which cryptocurrency market participants were able to exit their Terra stablecoin positions last Monday and Tuesday at close to par value, while retail holders of Terra and Luna have lost nearly all their money.


This is going to get uncomfortable, I suspect. People lost a lot of money. If some got preferential treatment, this will resound across the sector.
unique link to this extract

Miami’s mayor backed MiamiCoin. Then its price dropped 95% • Quartz

Scott Nover and Camille Squires:


On Feb. 2, the city of Miami cashed out its cryptocurrency MiamiCoin for the first time, depositing $5.25m into city coffers. Miami mayor Francis Suarez hailed it as a “historic moment” and predicted the cryptocurrency could one day even replace municipal taxes as the government’s primary source of funding.

MiamiCoin’s creator, an organization called CityCoins, has been no less enthusiastic, portraying the coin as a financial experiment that will empower citizens with a “community-driven revenue stream” while spurring new digital city services.

Miami is not the only city with big cryptocurrency dreams. CityCoins announced a similar cryptocurrency for New York in November 2021, and plans to release a coin for Austin, Texas, soon. Other cities have launched their own crypto ventures: Forth Worth, Texas, for example, will soon be running bitcoin mining rigs in city hall.

But only Miami’s mayor has thrown his full endorsement behind a CityCoin-branded cryptocurrency so far. After promoting MiamiCoin to residents and investors since its launch in August, the city of Miami received millions of dollars through its agreement with CityCoins.

Over the last nine months, however, MiamiCoin has lost nearly all of its value, falling about 95% from its September peak to just $0.0032 as of May 13. Its rapid descent has burned investors on the way down, muting the dreams of Miami’s city leaders, and possibly raising red flags for regulators now investigating cryptocurrency transactions.


New York got NYCCoin in November 2021, and its new mayor tweeted enthusiastically about it in January. Since then it’s fallen in value by 68%. Still time to get out.
unique link to this extract

Europe’s largest floating solar farm is ready to come online in Portugal • Electrek

Michelle Lewis:


The Alqueva reservoir is western Europe’s largest artificial lake, and the Alqueva Dam is on the Guadiana River, one of the longest in the Iberian Peninsula. The dam is in Alentejo, which is in southern Portugal, near the Spanish border.

The floating solar farm, the size of four soccer fields, is made up of 12,000 solar panels that will generate 7.5GWh annually and will also be paired with lithium batteries that can store 2 GWh. It will be able to power around 1,500 households. EDP, Portugal’s main utility company, built the floating solar farm.

The cool thing about floating solar farms on hydropower reservoirs is that they can be connected to existing links to the power grid. And as Reuters points out, “Excess power generated on sunny days can pump water up into the lake to be stored for use on cloudy days or at night.”

The Alqueva floating solar farm is furthering EDP’s plan to reach net zero by 2030. Renewables, including hydropower, now make up 78% of EDP’s 25.6 gigawatts (GW) of installed capacity.

EDP will expand the Alqueva floating solar farm because, last month, it secured the right to build a second, 70-MW installed capacity floating farm there.


unique link to this extract

• Why do social networks drive us a little mad?
• Why does angry content seem to dominate what we see?
• How much of a role do algorithms play in affecting what we see and do online?
• What can we do about it?
• Did Facebook have any inkling of what was coming in Myanmar in 2016?

Read Social Warming, my latest book, and find answers – and more.

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 )

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.