In cricket, where and when should a batsman look as the bowler runs up and then releases the ball? CC-licensed photo by WeLiveCricket.com 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.
On Friday, there’s another post due at the Social Warming Substack at about 0845 UK time.
A selection of 9 links for you. Not four or six. I’m @charlesarthur on Twitter. On Mastodon: https://newsie.social/@charlesarthur. Observations and links welcome.
How to blow up a timeline • Remains of the Day
Eugene Wei:
»
This past year, for the first time, I could see the end of the road for Twitter. Not in an abstract way; I felt its decline. Don’t misunderstand me; Twitter will persist in a deteriorated state, perhaps indefinitely. However, it’s already a pale shadow of what it was at its peak. The cool kids are no longer sitting over in bottle service knocking out banger tweets. Instead, the timeline is filled with more and more strangers the bouncer let in to shill their tweetstorms, many of them Twitter Verified accounts who paid the grand fee of $8 a month for the privilege. In the past year, so many random meetings I have with one-time Twitter junkies begin with a long sigh and then a question that is more lamentation than anything else: “How did Twitter get so bad?”
It’s sad, but it’s also a fascinating case study. The internet is still so young that it’s still momentous to see a social network of some scale and lifespan suddenly lose its vitality. The regime change to Elon and his brain trust and the drastic changes they’ve made constitute a natural experiment we don’t see often. Usually, social networks are killed off by something exogenous, usually another, newer social network. Twitter went out and bought Chekhov’s gun in the first act and use it to shoot itself in the foot in the third act. Zuckerberg can now extend his quip about Twitter being a clown car that fell into a gold mine.
«
Absolutely anything that Wei writes is obligatory reading. Read all this post. He feels things that others do, but then he puts it into words and you say “oh, that’s what I’ve been feeling”.
unique link to this extract
What does a batsman see? • Cricket Monthly
SB Tang:
»
[Australian Test batman Greg] Chappell realised that he had three ascending levels of mental concentration: awareness, fine focus and fierce focus. In order to conserve his finite quantum of mental energy, he would have to use fierce focus as little as possible, so that it was always available when he really needed it. When he walked out to bat, his concentration would be set at its lowest, power-saving level: awareness. He would mark his guard and look around the field, methodically counting all ten fielders until his gaze reached the face of the bowler standing at the top of his mark.
At that point, he would increase his level of concentration to fine focus. As the bowler ran in, he would gently and rhythmically tap his bat on the ground, keeping his central vision on the bowler’s face and his peripheral vision on the bowler’s body. He believed that a bowler’s facial expression and the bodily movements in his run-up and load-up offered the batsman valuable predictive clues as to what ball would be bowled. He would not look at the ball in the bowler’s hand as he ran in.
As the bowler jumped into his delivery stride, he would switch up his concentration to its maximum level – fierce focus – and shift his central vision the short distance from the bowler’s face to the window just above and next to his head from where he would release the ball. Once the ball appeared in that window, Chappell would watch the ball itself for the first time. He could see everything. He could see the seam of the ball and the shiny and rough side of the ball, even when he was facing a genuine fast bowler. Against spinners, he could see the ball spinning in the air as it travelled towards him. In the unlikely event that he failed to pick what delivery it was out of the hand, he could simply pick it in the air.
“There weren’t too many balls that I faced that I was unsure about,” Chappell tells the Cricket Monthly matter-of-factly. Because he was able to so quickly decipher where a ball was going to be, he was able to confidently move into position early to, if at all possible, play an attacking, run-scoring shot.
«
Fascinating account of how Chappell turned his mediocre batting around. If you’ve read Tim Gallwey’s book The Inner Game Of Tennis you’ll recognise a lot of what he then says.
unique link to this extract
CURSED HEIDI: AI-generated movie trailer • YouTube
I watched this, so I think you should. It’s like something generated by a field of magic mushrooms.
unique link to this extract
Apple purges predatory lending apps in India following scrutiny • TechCrunch
Manish Singh:
»
Apple removed several predatory lending apps from the App Store in India this week, days after users and media questioned the legitimacy of those services.
Pocket Kash, White Kash, Golden Kash and OK Rupee are among the apps that Apple pulled from the store this week. The apps offered fast-track lending to consumers in India, climbing to the top 20 of the finance list on the App Store in recent weeks. But they also levied outrageously superfluous charges, according to hundreds of user reviews.
The lenders also employed downright unethical tactics to get the borrowers to pay back.
“I borrowed an amount in a helpless situation and […] a day before repayment due date I got some messages with my pic and my contacts in my phone saying that repay your loan otherwise they will inform our contacts that you r not paying loan,” a user review from last month said.
The apps — whose developers had strange names and suspicious websites — were littered with hundreds of similar reviews, some sharing even more alarming threats that they allegedly received from the lenders.
«
Seems it was particularly the media making a noise to Apple which led to this.
unique link to this extract
AI-text detection tools are really easy to fool • MIT Technology Review
Rhiannon Williams:
»
Within weeks of ChatGPT’s launch, there were fears that students would be using the chatbot to spin up passable essays in seconds. In response to those fears, startups started making products that promise to spot whether text was written by a human or a machine.
The problem is that it’s relatively simple to trick these tools and avoid detection, according to new research that has not yet been peer reviewed.
Debora Weber-Wulff, a professor of media and computing at the University of Applied Sciences, HTW Berlin, worked with a group of researchers from a variety of universities to assess the ability of 14 tools, including Turnitin, GPT Zero, and Compilatio, to detect text written by OpenAI’s ChatGPT.
Most of these tools work by looking for hallmarks of AI-generated text, including repetition, and then calculating the likelihood that the text was generated by AI. But the team found that all those tested struggled to pick up ChatGPT-generated text that had been slightly rearranged by humans and obfuscated by a paraphrasing tool, suggesting that all students need to do is slightly adapt the essays the AI generates to get past the detectors.
“These tools don’t work,” says Weber-Wulff. “They don’t do what they say they do. They’re not detectors of AI.”
«
That noise you heard was the starting gun being fired on the arms race between the chatbots and the wannabe chatbot detectors. Chatbots well in the lead so far.
unique link to this extract
This is how the Netflix generation will watch F1 in the future • Forbes
Barry Collins:
»
Formula 1 fans may one day be able to watch races via an augmented reality headset, accessing live data feeds that will “make everybody a race engineer.”
Speaking ahead of the weekend’s British Grand Prix, Rob Smedley—the former Ferrari race engineer and now an F1 consultant—told me in a video interview that the sport is looking at new ways to engage the younger generation of fans that have flooded to F1 in recent years on the back of Netflix’s Drive to Survive series.
Smedley is part of the team that has delivered F1 Insights powered by AWS, a series of innovations that rely on cloud computing to deliver real-time data such as predicted pit-stop strategies or forecasts of forthcoming track battles, which are relayed live by race broadcasters.
Soon, Smedley believes that viewers wearing augmented reality headsets will be able to choose which data and video feeds they want to see, creating a virtual dashboard similar to that used by team race engineers.
Talking of Apple’s recently announced Vision Pro headset, Smedley described it as “really cool technology because you’re still in the room with everybody, but you’ve got this 4K screen in front of you. That, for me, is the future of sports watching.”
F1 and AWS are currently working on what Smedley described as “the second screen”, where a user-configurable dashboard of data is presented to viewers. That will intially be delivered via tablet devices that users can access alongside live race feeds, but Smedley said the project could eventually find its way into AR headsets such as Apple’s Vision Pro.
«
Sports is going to be such a huge thing for AR headsets once you can get the feedthrough. It’s odd, really, that Facebook/Meta hasn’t tried to focus more on it.
unique link to this extract
European court extends ‘right to be forgotten’ to news websites • Press Gazette
Charlotte Tobitt:
»
The European Court of Human Rights has been accused of approving the “rewriting of history” by backing the extension of the “right to be forgotten” from search engines to cover news websites more broadly.
The case, involving French-language Belgian newspaper Le Soir, last year saw an intervention from UK publishers Times Newspapers and Guardian News and Media alongside press freedom organisations as they argued forcing news websites to remove archive material, an “essential component of modern-day newsgathering and reporting”, would not be a “proportionate restriction on freedom of expression”.
However the Grand Chamber of the European Court of Human Rights, of which the UK is still a participating country despite Brexit, has now ruled in favour of a driver who had wanted to be anonymised in reporting of a deadly car crash for which he was responsible.
The original article was written in 1994 but went online in 2008 when Le Soir created an online version of its archives dating back to 1989. The driver, a doctor, first wrote to the newspaper’s owner asking for the article to be removed, or for him to at least be made anonymous, in 2010 – but his request was refused.
‘Right to be forgotten’ judge says decision does not ‘sacrifice press freedom’
Two years ago the ECHR rejected a free speech complaint from Le Soir publisher Patrick Hurbain who had argued that a Belgian court’s order to remove the name breached his rights under Article 10 (freedom of expression) of the European Convention on Human Rights, even when balanced with the right to be forgotten under Article 8 (privacy).«
I backed the original RTBF because the information still existed in newspaper and other archives. This, though, is wrong: it’s memory-holing the truth, and only the rich will really be able to afford to hassle the news organisations to make them expunge this. I rarely say this, but: the judges have got this completely wrong, and it’s good not to be in the EU so that this doesn’t apply to British publishers.
unique link to this extract
“We have built a giant treadmill that we can’t get off”: sci-fi prophet Ted Chiang on how to best think about about AI • Vanity Fair
Delia Cai talks to Chiang:
»
DC: How plugged in are you to the daily churn of tech news? I’m curious if you keep up with things like Marc Andreessen’s blog post about AI.
Ted Chiang: I am not, although I guess I’ll say I’m not super interested in what Marc Andreessen has to say. In general, I can’t say that I really keep up in any systematic fashion. But nowadays, you almost have to make a deliberate effort to avoid hearing about AI.
DC: Would you consider yourself to be an early adopter?
TC: Not of most technologies. I feel like being an early adopter requires a real commitment to constantly getting used to a new UI. I’m interested to see what is happening in technology, but in terms of my day-to-day work, I’m not looking for new software unless there’s an actual problem that I’m having. I wish I could still use a much older version of Word than I have to.
DC: How did your relationship with The New Yorker come about? The first couple of things you wrote for them back in 2016 and 2017 weren’t about tech per se. Who put it together that you should do those AI explainers?
TC: The first thing I wrote was the one about Chinese characters—they approached me, and I was totally flabbergasted. You know, I come from a science fiction background, and within the science fiction community, there’s a very clear border between science fiction and the literary establishment. It boggled my mind that anyone at The New Yorker had even heard of me.
There’s a piece from 2021, “Why Computers Won’t Make Themselves Smarter”, that is about some thoughts I had for a long time about the whole discourse around singularity and the explosion in machine intelligence, which I’m super, super skeptical of.. Earlier this year, I’d been reading about ChatGPT, trying to make sense of it for my own purposes, and around that time, [my editor] emailed me and said, “Do you have any thoughts on ChatGPT?” I was like, coincidentally, I do!
«
Chiang’s always worth reading. His short stories are fantastic.
unique link to this extract
Meta’s new Threads app raises potential privacy concerns over data sharing • Fox Business
Kelly O’Grady:
»
Meta’s new “Twitter killer” app Threads is the latest social media platform to burst onto the scene and is taking the industry by storm, with over 100 million users signing up less than a week after it launched.
But experts are warning of potential privacy concerns, particularly in the way Meta handles the data it collects from users when they subscribe to the new service. This includes sharing it with other platforms, including ones that may not have as strict data privacy protections, or that could even have servers in China.
Buried in the terms of service is a pledge to soon make Threads part of the “fediverse” – a decentralized network of servers that allows member social networks, like Mastodon, to communicate with each other. For example, a Threads user would be able to interact with a Mastodon user seamlessly, despite being on different platforms.
The upside is an online network that can be used without ever creating a profile or sharing personal data. The downside is Threads users with a public profile have already signed away that access.
Once the app is a part of the fediverse, Meta says: “Please be aware that you are directing us to deliver your information to services not controlled by Meta… Information sent to Third Party Services is no longer in Meta’s control and is subject to the terms and policies of those Third Party Services.”
«
Mastodon users: haha! Look, you sheeple, all your data might be in CHINA because it’s being spread around by Meta, because it’s in the fediverse, just like… Mastodon?
And – 100 million signups. That’s astonishing. It hasn’t been a week yet.
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
The Overspill: when there's more that I want to say 