Start Up No.2340: reading past the headlines, what Bluesky needs next, how people are using chatbots, suicide hold music, and more


Red squirrel or grey squirrel? An AI system is being used to distinguish the two, and help the red ones. CC-licensed photo by Shawn Nystrand 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. Distinguishable. I’m @charlesarthur on Twitter. On Threads: charles_arthur. On Mastodon: https://newsie.social/@charlesarthur. Observations and links welcome.


Social media users probably won’t read beyond this headline, researchers say • Penn State University

Ashley Wenners Herron:

»

Congratulations. Reading this far into the story is a feat not many will accomplish, especially if shared on Facebook, according to a team led by Penn State researchers. In an analysis of more than 35 million public posts containing links that were shared extensively on the social media platform between 2017 and 2020, the researchers found that around 75% of the shares were made without the posters clicking the link first. Of these, political content from both ends of the spectrum was shared without clicking more often than politically neutral content.

The findings, which the researchers said suggest that social media users tend to merely read headlines and blurbs rather than fully engage with core content, appeared on Nov. 19 in Nature Human Behavior. While the data were limited to Facebook, the researchers said the findings could likely map to other social media platforms and help explain why misinformation can spread so quickly online.

“It was a big surprise to find out that more than 75% of the time, the links shared on Facebook were shared without the user clicking through first,” said corresponding author S. Shyam Sundar, Evan Pugh University Professor and the James P. Jimirro Professor of Media Effects at Penn State. “I had assumed that if someone shared something, they read and thought about it, that they’re supporting or even championing the content. You might expect that maybe a few people would occasionally share content without thinking it through, but for most shares to be like this? That was a surprising, very scary finding.”

«

I feel that the good professor can’t have spent any actual time on social media, because if he had then he’d know for sure that people share articles willy-nilly, and the headline is the start and finish of their examination of it. Twitter introduced an element back in mid-2020 asking if you wanted to read the article first if you didn’t seem to have read it first. Hard to say whether it has really made a lot of difference, especially since links have been downgraded on X, and Musk doesn’t bother to do the most cursory examination of what he amplifies.
unique link to this extract


Twitter’s heir apparent isn’t X or Threads — it’s Bluesky • The Verge

Jay Peters:

»

Things haven’t been perfect for Bluesky; the platform has had some hiccups, seemingly due to the influx of people, and one major outage due to a cut fiber cable. But Meta has clearly been feeling the heat.

Last week, the day after Bluesky touted its 15 million total users, Threads boss Adam Mosseri shared that Threads had already surpassed 15 million signups for November alone. (CEO Mark Zuckerberg recently shared that the platform has 275 million monthly users.) And Meta has aggressively introduced some significant Threads changes, including its own version of custom feeds, a major change to make the For You feed show more from people you follow, and some big improvements to search.

Even if Bluesky is my preferred place right now, it’s still comparatively small to Threads and X. It’s unclear if people will stick with it over the long term, especially since Meta seems committed to making Threads its next billion-user social network. And X is still going to matter, given that many companies post updates on the platform and because owner Elon Musk is set to have enormous power in the Trump administration.

«

I think – if Bluesky does want to become an authoritative source – then it needs to add verification of users, and then perhaps to amplify them. But it’s still in its early phase; it hasn’t had its Miracle In The Hudson moment.
unique link to this extract


How WSJ readers use AI: from brainstorming to learning a new language and more • WSJ

Demetria Gallegos:

»

we asked Wall Street Journal readers: Tell us how you have used ChatGPT and other generative AI tools in your daily life, both personal and professional.

Here’s what they said.

• …I use Perplexity for understanding background to the news, for satisfying my curiosities about science, and sometimes to solve household maintenance issues. For example, I recently became interested in amateur radio astronomy. I researched what computing power I might need to pursue the hobby but got bogged down until I wrote one question to Perplexity AI. From its response I found I will need only slightly more processing power than I had expected. The bottom line for me is: For $20/month, I can save time and money by making a better-informed decision. I use it about 15 times a month.

• I tried using ChatGPT to suggest improvements on a watercolor I painted. First, it complimented the painting and then gave me six suggestions (all good) on how to make it even better. Among the tips, it wanted me to increase the contrast where the dock of my boatyard painting met the water, or where the building cast shadows. It also suggested I add a small foreground element such as a canoe or a few floating leaves to guide a viewer’s eye into the painting and add a sense of scale. Its final comment was “happy painting.”

•…My daughter and I were watching a History Channel program on the Roman Republic, a segment talking about the Roman legions. She’s a fan of a rather violent pseudo-historical series called “The Vikings,” and at the end of our program she turned to me and asked, “If the Romans and the Vikings had a war, who would win?” I said, ”Let’s ask ChatGPT.”

We got more than I bargained for—extended analysis about this hypothetical matchup, including specifics of battlefields and logistics. It analyzed the Romans’ respective strengths, including organization and discipline, training, sophistication of tactics, engineering and weapons. The Vikings were credited with their guerrilla tactics, naval skills, ferocity and skill in one-on-one combat.

ChatGPT concluded the Romans would likely win. My daughter was unconvinced her fearsome Vikings could be defeated.

«

unique link to this extract


Conservationists turn to AI in battle to save red squirrels • BBC News

Zoe Kleinman:

»

An artificial intelligence (AI) tool which has been trained to tell the difference between grey and red squirrels could be “an absolute game changer”, conservationists say.

The system, called Squirrel Agent, has been trained on thousands of images of the animals allowing it to tell them apart with 97% accuracy, its developer says.

It can then be used to automatically control access to squirrel feeders – with only reds being allowed into those containing food, and only greys into those where food has been replaced with contraceptive paste.

“It’s a real showcase of what AI can do,” said Emma McClenaghan, co-founder of Genysys Engine, which developed the tool. “It’s working in real time to do a task that we don’t have enough [human] volunteers to do.”

Squirrel Agent is currently being tested in sites around the UK in conjunction with five wildlife charities.
Genysys Engine hopes it will eventually be used much more widely not just with squirrels but with other species which would benefit from sophisticated digital monitoring.

Ian Glendinning, from Northern Red Squirrels – one of the conservation groups involved in the trial – told the BBC that, for the animals he is trying to protect, help from technology was urgently needed.

“We are in the bar of the last chance saloon, and the landlord has just called last orders,” he said.

«

Interesting idea. Grey squirrels are carriers of a virus that kills red squirrels, so maybe we also need the AI feeders to inject the red squirrels with a vaccine? Otherwise it’s hard to see things changing that much.

unique link to this extract


Hold on, the suicide counsellor will be with you presently • Radiolab

Simon Adler:

»

Two years ago, the United States did something amazing. In response to the mental health crisis the federal government launched 988 – a nationwide, easy to remember phone number that anyone can call anytime and talk to a counselor. It was 911 but for mental health and they hoped that it would save lives. However, if you call 988 today the first thing you hear isn’t a sympathetic counselor. What you hear is hold music.

Today, the story of the highest stakes hold music in the universe, the three men who created suicide prevention and the two women trying to fix it. 

«

You can listen to the podcast, or read the transcript.

Initially, the hold music was “snazzy jazz music”. Really you have to listen to it to understand the nuances.

(And no, it wasn’t the M*A*S*H* theme.)
unique link to this extract


WhatsApp gains voice message transcripts • MacRumors

Juli Clover:

»

Popular Meta-owned messaging app WhatsApp has announced a new transcription feature that’s designed to provide you with a transcript of a voice message received from a friend or family member.

WhatsApp says that voice message transcripts are designed for instances when you’re in a loud place and can’t stop to listen to a traditional voice message. Transcripts are created on-device, are end-to-end encrypted, and aren’t shared with WhatsApp.

WhatsApp users can go to Settings > Chats > Voice Message Transcripts to turn transcriptions on and off and to select a preferred transcript language. A voice message can be transcribed by long pressing on it and tapping on the “transcribe” option.

The voice message transcript feature has been available to WhatsApp beta testers for months now, but it is now ready to start rolling out to all users.

«

Smart idea. Transcription is going to become something that within a year or two is utterly routine, after decades and decades when it seemed just in reach. (In passing, my version hasn’t yet got this option.)
unique link to this extract


Sweden’s Northvolt files for bankruptcy, in blow to Europe’s EV ambitions • Reuters

Dietrich Knauth, Marie Mannes and Terje Solsvik:

»

Northvolt, the Swedish maker of battery cells for electric vehicles, said on Thursday it has filed for Chapter 11 bankruptcy protection in the US, dealing a blow to Europe’s hopes that its most developed battery player would reduce Western car makers’ reliance on Chinese rivals.

Northvolt said it has only enough cash to support operations for about a week and said it has secured $100m in new financing for the bankruptcy process. It said operations will continue as normal during the bankruptcy.

“Northvolt’s liquidity picture has become dire,” the company said in its Chapter 11 petition, filed in US Bankruptcy Court in Houston. The company, which has operations in California, has about $30m of cash, which can support its operations for only about a week. It has $5.8bn in debts.

Northvolt, which employs around 6,600 staff across seven countries, said it expects to complete the restructuring by the first quarter of 2025.

Northvolt transformed in a matter of months from Europe’s best shot at a homegrown electric-vehicle battery champion to a company struggling to stay afloat by slimming down, hobbled by production problems, the loss of a major customer and a lack of funding.

Europe has been hoping that Northvolt would reduce Western car makers’ reliance on Chinese rivals such as battery maker CATL and EV and battery maker BYD.

«

They seem to be blaming both lack of demand and missed targets, which seems contradictory.
unique link to this extract


Kate Nash says her OnlyFans photos will earn more than tour • BBC News

Ian Youngs:

»

Nash, who released her fifth studio album in June, also told fans on Instagram: “No need to stream my music, I’m good for the 0.003 of a penny per stream thanks.”

Last month, fellow singer Lily Allen revealed she makes more money by selling pictures of her feet on OnlyFans than she does from Spotify streams.

Meanwhile, in recent months, acts including Rachel Chinouriri, Ratboy and The Duke Spirit singer Liela Moss have all cancelled tours, blaming the costs.

Nash highlighted a survey from recording and rehearsal studio network Pirate, which said most artists have not seen an increase in gig fees in recent years despite a rise in ticket prices.

“Festival prices and ticket prices have gone up drastically, but the musicians’ wage hasn’t,” she said. “So you might be playing a venue that you’ve played multiple times and you can sell it out, [but] you’re getting the same fee that you did 10 years ago, probably. But all the other costs have gone up.”

Some corporations make big profits from music, as do a “select few” artists, she said. “But the majority are losing money, and we are also creating an environment where the industry is saying, we don’t want diversity in music, because we don’t want working class people to be able to afford to do this.”

Musicians could follow a lead from people who earn a living from selling sexual content on sites like OnlyFans, she suggested. “You’ve got all this control, and you’re deciding what you want to do and how you want to do it, and people want to pay you for it.

“We just haven’t taught any of those lessons to anyone with music and art – that art is so valuable and so worthwhile in our lives and so meaningful. We’re totally happy to devalue it.

“Where can we learn from the sex workers? Maybe we can learn something from this industry. How do we get empowered as artists and take a bit more control?”

«

That point about the cut going to artists not rising is where the Competition & Markets Authority should be getting involved. Ticketing companies are raking it in.
unique link to this extract


Source: Google has canceled the Pixel Tablet 2 – but not the Tab 3? • Android Authority

Mishaal Rahman:

»

Android Authority has learned that Google has cancelled the Pixel Tablet 2, the presumed name of Google’s second-generation Pixel Tablet. This is disappointing for Pixel fans who were waiting for Google to refresh its first-generation Pixel Tablet with a newer chipset, a better camera, and, more importantly, an official keyboard accessory.

It’s also surprising to hear because it might suggest that Google is giving up on its tablet ambitions entirely, considering a separate report published last week claimed that Google is also killing the Pixel Tablet 3. However, we have reason to believe that the device cited in yesterday’s report is actually the Pixel Tablet 2, and not the third-generation tablet after all.

Last week, I shared what I learned about the Pixel Tablet 2 from a source within Google. I deemed this source to be very credible given my past history with them as well as the fact that they were able to share unreleased images of the device with me (which I obviously did not publish to protect their identity). After the publication of this article, however, I learned from my source that Google had decided to cancel its plans to release the device, citing concerns that the company would lose money on it.

On Wednesday, Android Headlines published an article claiming that Google has canceled development of the Pixel Tablet 3 and not the Pixel Tablet 2 that Google was actively working on. I found this report to be strange, considering that Google has almost assuredly not started working on the Pixel Tablet 3 while the Pixel Tablet 2 is still in an early stage of the product development lifecycle. I had heard that the device that had been cancelled was the very device I had reported on last week, and that device is almost certainly not the Pixel Tablet 3.

«

Google’s hate-hate relationship with tablets continues: aside from the Nexus 7in tablet, in 2012, its first effort – which was a success – it has always struggled to find a reason to have tablets in its lineup. There is talk of “merging-but-not-merging” ChromeOS and Android, though who knows how that’s going to go.
unique link to this extract


First case of bird flu infection in a US child confirmed in California, officials say • Associated Press via PBS News

Mike Stobbe and JoNel Aleccia:

»

Health officials on Friday confirmed bird flu in a California child — the first reported case in a US minor.

The child had mild symptoms, was treated with antiviral medication and is recovering, the Centers for Disease Control and Prevention said in announcing the test results. State officials have said the child attends day care and lives in Alameda County, which includes Oakland and surrounding communities, but released no other details.

The infection brings the reported number of US bird flu cases this year to 55, including 29 in California, the CDC said. Most were farmworkers who tested positive with mild symptoms.

One exception was an adult in Missouri who did not work at a farm and had no known contact with an infected animal. It remains a mystery how that person was infected — health officials have said there was no evidence of it spreading between people.

A British Columbia teen also was recently hospitalized with bird flu, Canadian officials have said.

H5N1 bird flu has been spreading widely in the U.S. among wild birds, poultry and a number of other animals over the last few years.

It began spreading in US dairy cattle in March. California has become the center of that outbreak, with 402 infected herds detected there since August. That’s 65% of the 616 herds confirmed with the virus in 15 states.

«

What is faintly interesting is that the story says “US child” rather than “American child”, which to me subtly implies that this is the child of an immigrant worker. The CDC statement is also a little cagey, and doesn’t make it absolutely clear whether the child worked on a farm, though the strong implication is that they didn’t. In which case it’s a human carrier transmitting it.
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 comment

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