Start Up No.1432: Facebook labels don’t stop Trump’s lies spreading, when will Twitter ban him?, Apple’s M1 machines benchmarked, and more


Unable to cope with being cut off from manufacturers and vendors, Huawei has sold its Honor low-cost smartphone business to a Shenzhen company. CC-licensed photo by Marco Verch 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 12 links for you. Not laughed out of court. I’m @charlesarthur on Twitter. Observations and links welcome.

Facebook labels on Trump’s lies do little to stop spread • Buzzfeed News

Craig Silverman and Ryan Mac:

»

The labels Facebook has been putting on false election posts from President Donald Trump have failed to slow their spread across the platform, according to internal data seen by BuzzFeed News.

In the aftermath of the 2020 US presidential election, Trump has repeatedly spread false information questioning President-elect Joe Biden’s victory — and been rewarded with massive engagement on Facebook. The company has attempted to temper this by adding labels to the false claims directing people to accurate information about the election and its results.

But this has done little to prevent Trump’s false claims from going viral on Facebook, according to discussions on internal company discussion boards. After an employee asked last week whether Facebook has any data about the effectiveness of the labels, a data scientist revealed that the labels — referred to as “informs” internally — do very little to reduce them from being shared.

”We have evidence that applying these informs to posts decreases their reshares by ~8%,” the data scientists said. “However given that Trump has SO many shares on any given post, the decrease is not going to change shares by orders of magnitude.”

The data scientist noted that adding the labels was not expected to reduce the spread of false content. Instead, they are used “to provide factual information in context to the post.”

“Ahead of this election we developed informational labels, which we applied to candidate posts with a goal of connecting people with reliable sources about the election,” Facebook spokesperson Liz Bourgeois said in a statement, adding that labels were “just one piece of our larger election integrity efforts.”

«

Letting destabilising lies spread effectively unchecked: it’s hard to portray Facebook’s effect positively, really.
unique link to this extract


Statement • Huawei

»

Huawei’s consumer business has been under tremendous pressure as of late. This has been due to a persistent unavailability of technical elements needed for our mobile phone business. Huawei Investment & Holding Co., Ltd. has thus decided to sell all of its Honor business assets to Shenzhen Zhixin New Information Technology Co., Ltd. This sale will help Honor’s channel sellers and suppliers make it through this difficult time.

Once the sale is complete, Huawei will not hold any shares or be involved in any business management or decision-making activities in the new Honor company.

This move has been made by Honor’s industry chain to ensure its own survival. Over 30 agents and dealers of the Honor brand first proposed this acquisition.

Since its creation in 2013, the Honor brand has focused on the youth market by offering phones in the low- to mid-end price range. During these past seven years, Honor has developed into a smartphone brand that ships over 70 million units annually.

«

No price put on this. In theory, Huawei’s smartphone business should be worth billions and billions, but I wonder if the Chinese government encouraged a bank to help Shenzhen Zhixin New Information Technology with a friendly-terms loan. (There’s still the high-end smartphone business which it appears to have retained.)

And whether, if Joe Biden’s administration makes some sort of deal with China, whether Huawei will take its business back just as abruptly.
unique link to this extract


Will Twitter ban Trump? • The Atlantic

Kaitlyn Tiffany:

»

This year, Twitter did start moderating the president—and misinformation in general—meaningfully for the first time. In the spring, when Trump spread blatant disinformation about the coronavirus, the company slapped warning labels on his tweets. (It also briefly suspended his son Donald Trump Jr. for sharing a video that claims hydroxychloroquine is a cure for COVID-19.) Throughout the summer and fall, President Trump’s lies about mail-in voting and election fraud have been appended with notes about the disputed claims and links to real information. Since November 4, the president has tweeted (or retweeted himself) more than 120 times. So far, about 40 of these tweets come with a warning label. (This is not counting the claims of election fraud he has retweeted from other people.)

Now he is two months away from losing exemption from Twitter’s rules—theoretically going back to the same treatment as anyone else. “[The world-leader] policy framework applies to current world leaders and candidates for office, and not private citizens when they no longer hold these positions,” a Twitter spokesperson confirmed in a statement. Twitter has spent the past year putting checks on the president’s speech, adding friction to the process by which conspiracy theories spread, and labeling false information for what it is. But it hasn’t yet gone nuclear. The company is in a bind: Banning Trump after he leaves office would be interpreted as an aggressive political act by much of the right. It might drive him and his followers to other, more insular parts of the internet, where delusions and lies would go unchecked by the mainstream. But allowing him to keep spreading dangerous misinformation would be hypocritical, and, frankly, bad for the company’s image.

«

It’s only a matter of time. It will happen gradually – there will be a limited suspension for a tweet that must be deleted, and eventually something will tip it over. Yet Trump won’t want to lose his bully pulpit, so he might try to be careful. Twitter though is going to be aching for some reason to end the embarrassment.
unique link to this extract


Facebook didn’t fully enforce call to arms rule for events • Buzzfeed News

Ryan Mac and Craig Silverman:

»

In August, following a Facebook event at which two protesters were shot and killed in Kenosha, Wisconsin, Mark Zuckerberg called the company’s failure to take down the event page asking militant attendees to bring weapons “an operational mistake.” There had been a new policy established earlier that month “to restrict” the ability of right-wing militants to post or organize in groups, Facebook’s CEO said, and under that rule, the event page should have been removed.

BuzzFeed News has learned, however, that Facebook also failed to enforce a separate year-old call to arms policy that specifically prohibited event pages from encouraging people to bring weapons to intimidate and harass vulnerable individuals. Three sources familiar with the company’s content moderation practices told BuzzFeed News that Facebook had not instructed third-party content moderators to enforce a part of its call to arms policy that was first established in June 2019.

“What we learned after Kenosha is that Facebook’s call to arms policy didn’t just fail,” said Farhana Khera, the executive director of Muslim Advocates, a civil rights organization that pressured Facebook to create the rule. “It was never designed to function properly in the first place.”

«

unique link to this extract


Primary energy vs final energy: why replacing fossil fuels won’t be so hard • Bloomberg

Akshat Rathi:

»

The average efficiency of coal power plants globally is about 33%, according to the World Coal Association. That is, only a third of the energy stored in the black lumps is converted to electricity. Many modern internal combustion engine cars have an efficiency of about 20%, transforming only a fifth of gasoline’s energy into motion.

Why should something as dry and technical as energy-measurement warrant attention? Because misunderstanding it dramatically overplays the difficulty of meeting global clean-energy goals.

Let’s first define the terms. “Primary energy” is the measure of energy as found in nature, say blocks of coal or crude oil. “Final energy” is what’s available for us to use in the form of gasoline or electricity. And “useful energy” is the fraction that’s converted to, for example, move a car or light up a room.

Even though fossil fuels meet roughly 80% of the world’s primary energy demand, they are responsible for only 60% of its useful energy, according to BNEF. Put another way, 20% of the world’s primary energy demand today is met by non-fossil sources—and those sources are responsible for 40% of the world’s useful energy.

…As the world consumes more energy from renewables in the form of electricity, it manages similar economic output while consuming less primary energy. Generating electricity from solar and wind is highly efficient, and motors inside electric vehicles convert more than 80% of the energy stored in batteries into motion.

«

The maths seems reassuring, yet we never seem to be any closer to changing things.
unique link to this extract


Yeah, Apple’s M1 MacBook Pro is powerful, but it’s the battery life that will blow you away • TechCrunch

Matthew Panzarino:

»

The M1 MacBook Pro runs smoothly, launching apps so quickly that they’re often open before your cursor leaves your dock. 

Video editing and rendering is super performant, only falling behind older machines when it leverages the GPU heavily. And even then only with powerful dedicated cards like the 5500M or VEGA II. 

Compiling projects like WebKit produce better build times than nearly any machine (hell the M1 Mac Mini beats the Mac Pro by a few seconds). And it does it while using a fraction of the power. 

This thing works like an iPad. That’s the best way I can describe it succinctly. One illustration I have been using to describe what this will feel like to a user of current MacBooks is that of chronic pain. If you’ve ever dealt with ongoing pain from a condition or injury, and then had it be alleviated by medication, therapy or surgery, you know how the sudden relief feels. You’ve been carrying the load so long you didn’t know how heavy it was. That’s what moving to this M1 MacBook feels like after using other Macs. 

Every click is more responsive. Every interaction is immediate. It feels like an iOS device in all the best ways. 

«

He says the battery life is a minimum of two to three times that of previous Mac portables. And they were pretty good.
unique link to this extract


1% of people cause half of global aviation emissions – study • The Guardian

Damian Carrington:

»

Frequent-flying “‘super emitters” who represent just 1% of the world’s population caused half of aviation’s carbon emissions in 2018, according to a study.

Airlines produced a billion tonnes of CO2 and benefited from a $100bn (£75bn) subsidy by not paying for the climate damage they caused, the researchers estimated. The analysis draws together data to give the clearest global picture of the impact of frequent fliers.

Only 11% of the world’s population took a flight in 2018 and 4% flew abroad. US air passengers have by far the biggest carbon footprint among rich countries. Its aviation emissions are bigger than the next 10 countries combined, including the UK, Japan, Germany and Australia, the study reports.

The researchers said the study showed that an elite group enjoying frequent flights had a big impact on the climate crisis that affected everyone.

They said the 50% drop in passenger numbers in 2020 during the coronavirus pandemic should be an opportunity to make the aviation industry fairer and more sustainable. This could be done by putting green conditions on the huge bailouts governments were giving the industry, as had happened in France.

«

These frequent flyers aren’t called “pilots” by any chance, are they? I mean, they’re the source of the trouble.
unique link to this extract


The rise and fall of Getting Things Done • The New Yorker

Cal Newport:

»

the productivity pr0n movement continued to thrive because the overload culture that had inspired it continued to worsen. G.T.D. was joined by numerous other attempts to tame excessive work obligations, from the bullet-journal method, to the explosion in smartphone-based productivity apps, to my own contribution to the movement, a call to emphasize “deep” work over “shallow.” But none of these responses solved the underlying problem.

The knowledge sector’s insistence that productivity is a personal issue seems to have created a so-called “tragedy of the commons” scenario, in which individuals making reasonable decisions for themselves insure a negative group outcome. An office worker’s life is dramatically easier, in the moment, if she can send messages that demand immediate responses from her colleagues, or disseminate requests and tasks to others in an ad-hoc manner. But the cumulative effect of such constant, unstructured communication is cognitively harmful: on the receiving end, the deluge of information and demands makes work unmanageable.

There’s little that any one individual can do to fix the problem. A worker might send fewer e-mail requests to others, and become more structured about her work, but she’ll still receive requests from everyone else; meanwhile, if she decides to decrease the amount of time that she spends engaging with this harried digital din, she slows down other people’s work, creating frustration.

In this context, the shortcomings of personal-productivity systems like G.T.D. become clear. They don’t directly address the fundamental problem: the insidiously haphazard way that work unfolds at the organizational level. They only help individuals cope with its effects. A highly optimized implementation of G.T.D. might have helped [Merlin] Mann organize the hundreds of tasks that arrived haphazardly in his in-box daily, but it could do nothing to reduce the quantity of these requests.

«

I often tell people that your email inbox is a to-do list written by someone else, and that you should treat it with the disdain that that implies. (But there’s an ever-burgeoning market for to-do list apps.)
unique link to this extract


Apple hits back at European activist complaints against tracking tool • Reuters

Kirsti Knolle:

»

Noyb’s complaints were brought against Apple’s use of a tracking code, known as the Identifier for Advertisers (IDFA), that is automatically generated on every iPhone when it is set up.

The code, stored on the device, makes it possible to track a user’s online behaviour and consumption preferences – vital in allowing companies to send targeted adverts.

“Apple places codes that are comparable to a cookie in its phones without any consent by the user. This is a clear breach of European Union privacy laws,” Noyb lawyer Stefano Rossetti said.

Rossetti referred to the EU’s e-Privacy Directive, which requires a user’s consent before installation and using such information.

Apple said in response that it “does not access or use the IDFA on a user’s device for any purpose”.

It said its aim was to protect the privacy of its users and that the latest release of its iOS 14 operating system gave users greater control over whether apps could link with third parties for the purposes of targeted advertising.

The Californian tech giant said in September it would delay plans to launch iOS 14 [with IDFA blocking] until early next year.

Apple accounts for one in every four smartphones sold in Europe, according to Counterpoint Research.

«

unique link to this extract


Mark Zuckerberg said banning Steve Bannon from the platform for advocating for the beheading of Dr. Anthony Fauci and FBI director Christopher Wray is ‘not what our policies would suggest’ • Yahoo

»

Facebook CEO Mark Zuckerberg reiterated during Tuesday’s virtual congressional hearing that the firm will not suspend Steve Bannon’s account following his call for Dr. Anthony Fauci to be beheaded in a November 5 video.

In the hearing, Zuckerberg acknowledged that the video did violate Facebook’s policies, which is why the firm took the video down. But when Sen. Richard Blumenthal (D-CT) asked if Facebook would suspend the former Trump advisor’s account, Zuckerberg said the company would not.

“Senator, no, that’s not what our policies would suggest that we should do in this case,” Zuckerberg said in Tuesday’s hearing.

Bannon went after Fauci and FBI director Christopher Wray in a video of his podcast on November 5, which he recirculated on Facebook, as well as on Twitter and YouTube. He slammed the two for disagreeing with President Donald Trump and said “I’d put their heads on pikes” outside of the White House to warn other federal officials.

Facebook removed the video, but Zuckerberg then reportedly told employees in an all-hands meeting [last week] that Bannon hadn’t violated enough of the company’s policies to be booted from the platform. 

«

Three strikes and you’re.. not out. Four. OK, five. Well, look, there’s definitely a line and don’t cross it, OK? We’ll tell you when you do. (Twitter banned Bannon permanently for the same comments.)
unique link to this extract


A court ruling in Austria could censor the internet worldwide • Slate

Jennifer Daskal:

»

A little more than a year ago, I wrote with concern about the risk that a single EU court within single EU member state would become the censor for the world. That fear has now become reality. In a ruling in September, first reported on Thursday, the Austrian Supreme Court ordered, pursuant to local defamation rules, that Facebook remove a post insulting a former Green Party leader, keep equivalent posts off its site, and do so on a global scale.*

The case started with an April 2016 Facebook post, in which a user shared an article featuring a photo of Eva Glawischnig-Piesczek, then-chair of Austria’s Green Party, along with commentary labeling her a “lousy traitor,” “corrupt oaf,” and member of a “fascist party,” apparently in response to her immigration policies. This is core, protected speech in the United States. But it was deemed defamation under Austrian law. And in a series of rulings, Austrian courts ordered that Facebook take down and keep off any such post, and do so around the world.

Facebook complied, but only in part.  Employing what is known as geoblocking, it made the particular post that had been identified inaccessible to users within Austria. But it objected both to the global reach of the order and to the obligation to look for and keep other, equivalent posts off their site. And it argued that the order violated the applicable EU’s e-Commerce Directive, which prohibits EU member states from imposing general monitoring obligations on tech companies like Facebook.

In an October 2019 ruling, the European Court of Justice sided with Austria—concluding that the e-Commerce Directive did not stand in the Austrian court’s way (a ruling that is hard to reconcile with the text of the directive, for reasons I won’t rehash now). The CJEU ruling set two limits, albeit minimal, on EU member state courts. First, the obligation to look for and take down equivalent content must be issued with sufficient specificity to eliminate the need for independent provider assessment of that content—a criteria that, as I and others have argued previously is premised on a largely misguided assessment of the ability and precision of filtering tools. Second, any worldwide injunction must comport with “relevant international law.”

«

International law is getting very, very confusing but the emerging reality is that the only options are not to operate in certain countries, or comply with what they demand. Facebook’s success bites it again.
unique link to this extract


FCPX performance on M1 chip • MacRumors Forums

»

This is from a photographer in Australia. He got the device today and posted this on Weibo in Chinese. I guess I’ll translate a bit.

Exporting H.264 Sony 10 bit 422 footage with just one rec709 lut takes:

– 11 minutes and 30 seconds on his iMac Pro with Vega 56 and 128gigs of ram.
– 10 minutes and 20 seconds on the new MacBook Pro with 8gigs of ram took. Wow.

«

(FCPX = Apple’s Final Cut Pro X, used by video editors all the time.) There is more discussion and more benchmarks in the thread, but this suggests that even with minimal RAM the M1 can leave Intel chips in the dust. If word gets around fast enough (and among pro users it surely will), Apple might have trouble selling its Intel-based high-end machines because people will see that the lower-end ones smoke them already, and will hold out for the M*-based versions. The M* Mac Pro will be quite something, on this basis.

There’s also this Twitter thread comparing a compiler run on a M1 MacBook Air to a 16in (Intel) MacBook Pro. Developers are going to want to have their tools running on it, but might be happy with x86 emulation.

Meanwhile Anandtech have looked at the M1 Mac mini and conclude that Apple “hit it out of the park”, as it now beats Intel and is level with AMD’s best.
unique link to this extract


Errata, corrigenda and ai no corrida: Yes, the former president of the US is actually Barack Obama, not Barack Obaba. Thanks to all those who pointed out the error.

1 thought on “Start Up No.1432: Facebook labels don’t stop Trump’s lies spreading, when will Twitter ban him?, Apple’s M1 machines benchmarked, and more

  1. I wonder if there’s a prediction market / betting pool for the day that Twitter bans Trump? I suspect they will do it sooner rather than later (12:01 pm Jan 20 is NOT out of the question). They don’t want him playing President-In-Exile, that’s not worth any “engagement”. I don’t think they’re going to play around with individual quibbles or time limits. I’d say that if he doesn’t give them an immediate cause himself, they’ll just interpret something he said with maximum uncharitability and be done with it. It’s not like there’s some sort of appeal to anyone who can overrule them. Saying something like Trump dog-whistled a threat against the life of President Biden would work fine.

    In fact, is there anything in their policies that indicate that they use a model like if a Member of Parliament makes a privileged statement, they can’t be prosecuted for those even when they are no longer an MP, as opposed to one where you can’t get prosecuted during the time you’re in office, but the moment you lose office, it’s open season _for past incidents_? That is, according to their rules, are the policy-violating _tweets_ privileged (forever), or is the person just temporary privileged against banning for those tweets for a while? If Trump went on a horrible Twitter rampage at 11am Jan 20 (which could really happen!), would Twitter consider itself policy-bound not to ban for those tweets after noon Jan 20, when Trump is no longer President?

Leave a Reply

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

WordPress.com Logo

You are commenting using your WordPress.com 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.